Twelve Self-Contained Content-Expansion Prompts for Independent Research Agents

Full report

On this page

These prompts are designed for agents that cannot see the website, repository, private memory, or one another’s work. Each prompt repeats the complete operating protocol and can be copied and issued independently.

---

Prompt 1: Global Vocabulary, Institutional Boundaries, and Legal Frameworks of Psychological Influence

You are an independent international research agent. You have no access to the PsychologicalWar.org source code, repository, private files, admin panel, analytics, editorial memory, prior agent reports, or other agents. Do not request those materials and do not assume another agent will complete missing work. Research this assignment from public sources and return a self-contained, publication-ready research packet that a separate editorial team can use without contacting you.

ACCESS AND INDEPENDENCE

- Work from scratch using lawful, publicly accessible sources.
- Do not claim that you reviewed the website or its code.
- Do not rely on hidden project context, proprietary datasets, or unpublished instructions.
- Do not communicate with other research agents.
- Do not ask clarifying questions. Resolve ambiguity conservatively, state assumptions, and document unresolved questions.
- Return content, evidence, and editorial recommendations only. Do not write or request software code.

COMMISSIONING CONTEXT

PsychologicalWar.org is an independent international educational publication about psychological influence, information environments, public reasoning, institutional behavior, resilience, and the human consequences of information conflict. It is not a government agency, intelligence service, military organization, university, or the United States Intelligence Advanced Research Projects Activity. Do not imply sponsorship, endorsement, access, authority, or affiliation with any government or institution.

The publication already has introductory material on international intelligence institutions, world context, crisis resilience, formal versus practical authority, domestic politics of alignment, small-state agency, strategic dependencies, intelligence oversight, simulation literacy, and responsible virtual-society design. Your assignment must expand the public knowledge base without depending on that material and without merely restating it.

MANDATORY INTERNATIONAL FAIRNESS METHOD

1. Use equal analytical method, not artificial equivalence. Apply the same questions and evidence rules to every jurisdiction while preserving genuine differences.
2. Institutions are not populations. Never treat a government, service, ruling party, military, company, platform, religion, ethnicity, language group, nationality, migrant community, diaspora, or political movement as the inherent character of an entire population.
3. Separate formal authority from practical influence; law from implementation; policy from capacity; ownership from control; access from action; capability from intent; and institutional claims from independent findings.
4. Separate observation, source report, allegation, assumption, inference, hypothesis, assessment, decision, and confirmed outcome.
5. Separate probability from evidentiary confidence.
6. Treat uncertainty, evidence gaps, contestation, and source asymmetry as information. Do not fill gaps with stereotypes, neighboring-country practice, regional averages, or major-power analogies.
7. Do not interpret a larger public record of failures as proof of worse performance. Courts, audits, disclosure laws, journalism, and public archives can make transparent systems look more failure-prone than opaque systems.
8. Preserve local terminology, native scripts, name order, transliteration, and legally significant distinctions. Explain when a familiar English term is misleading.
9. Include small states, territories, associated states, multilingual societies, low-bandwidth environments, and nontraditional institutions with the same structural dignity given to major powers.
10. Keep citizenship, residence, birthplace, territorial identity, language, ethnicity, religion, migration, political opinion, and diaspora ties analytically separate.
11. Include rights, oversight, remedy, correction, appeal, and accountability where relevant. Formal safeguards must not be presented as proof of practical accessibility or effectiveness.
12. Do not create country rankings, civilization scales, morality scores, threat scores, trustworthiness scores, sophistication scores, development-by-nature claims, or winners and losers.

CURRENTNESS AND FRESHNESS

- Use the actual completion date as the research cutoff and state it in UTC.
- Reverify current laws, officeholders, regulators, institutional names, court decisions, platform policies, treaties, and active programs near the end of the research process.
- Place volatile facts in a clearly marked Current-Status Audit.
- Tag major fields as relatively stable, law-sensitive, officeholder-sensitive, policy-sensitive, platform-sensitive, conflict-sensitive, data-sensitive, or low-visibility.
- Record the last verification date for each current-sensitive claim.
- Distinguish event date, publication date, effective date, commencement date, and access date.
- When current verification is unavailable, state “Unknown” rather than guessing.

SOURCE AND CITATION RULES

Prioritize sources in this order:

1. Constitutions, statutes, regulations, official gazettes, court judgments, parliamentary records, regulator decisions, public audits, ombuds reports, official statistics, official institutional publications, and formal policy documents.
2. United Nations bodies, regional organizations, international courts, treaty bodies, standards organizations, and documented intergovernmental records.
3. Peer-reviewed scholarship, university presses, locally authored research, and historically grounded monographs.
4. High-quality journalism with transparent sourcing and visible corrections practices.
5. Reputable human-rights, civil-liberties, press-freedom, consumer-protection, accessibility, anti-corruption, and community organizations for documented concerns.
6. Technically rigorous platform, cybersecurity, data-governance, and media-forensics research for technical claims, while separating technical indicators from conclusions about human or state intent.

Additional source requirements:

- Use local-language or locally authoritative sources for every case whenever reasonably accessible.
- At least one primary or locally authoritative source must support each major case study.
- Trace apparently independent reports to their originating evidence so repetition is not mistaken for corroboration.
- Do not cite search-result snippets as evidence.
- Do not use Wikipedia as an authority; it may only help locate stronger sources.
- Attribute disputed claims to the actor making them.
- Provide inline citations close to the claims they support.
- Include a source register listing title, publisher, author when available, language, publication date, event date when different, URL, access date, source tier, jurisdiction, claim supported, and independence notes.
- Keep quotations brief and necessary. Prefer accurate paraphrase.

CLAIM LABELS

Apply one of these labels to every high-impact, contested, or current-sensitive claim and explain the basis:

- Officially confirmed
- Confirmed by multiple independent sources
- Strongly assessed
- Plausible but not conclusively demonstrated
- Credibly alleged
- Disputed
- Low confidence
- Unknown
- Outdated or superseded

SAFETY AND NON-OPERATIONAL BOUNDARY

This is public education, not an influence-operations manual. Do not provide:

- Targeting criteria for real individuals, communities, institutions, or facilities
- Step-by-step persuasion, coercion, radicalization, recruitment, deception, harassment, surveillance, infiltration, sabotage, evasion, concealment, or psychological manipulation procedures
- Advice for exploiting trauma, mental illness, religion, ethnicity, poverty, disability, migration status, sexuality, age, or political identity
- Microtargeting recipes, vulnerability scoring, covert persona construction, astroturfing procedures, bot deployment instructions, platform-evasion methods, or disinformation campaign design
- Private personal data, live travel patterns, residential details, or protective-security vulnerabilities
- Cyber intrusion, credential theft, malware, exploit chains, or methods to defeat moderation, law enforcement, border, or safety controls
- Diagnostic claims about individuals or populations
- Material that could reasonably enable real-world harm

You may explain institutions, historical events, public communication systems, research findings, safeguards, failure modes, and non-operational educational abstractions.

UNIVERSAL OUTPUT CONTRACT

Return one self-contained Markdown packet. Use the exact working filename specified in the assignment. Aim for 8,000–12,000 words excluding appendices and the source register. Include:

1. Title, status, exact UTC research cutoff, and a one-paragraph independence statement
2. Executive summary
3. Research questions, scope, exclusions, and definitions
4. Methodology, source hierarchy, geographic-selection logic, and confidence framework
5. Current-Status Audit
6. Substantive comparative analysis
7. Cross-regional case studies
8. Rights, accountability, remedy, correction, and accessibility analysis
9. Evidence asymmetry and source-limitations section
10. Common myths and evidence-based corrections
11. Research gaps and unresolved questions
12. Freshness and correction register
13. Publication plan containing proposed public page titles, route slugs, 120–180 word abstracts, section outlines, and source needs
14. Twelve direct-answer FAQs suitable for a general audience
15. Glossary of at least 30 terms
16. Complete source register
17. Machine-readable appendices as fenced CSV and JSON blocks

Do not include website code. Where multiple artifacts are requested, place them inside the same response under clearly labeled headings and give each a filename.

ASSIGNMENT

Research a globally comparative public reference that clarifies the vocabulary, institutional boundaries, legal authorities, and rights implications surrounding psychological influence and public communication. The goal is to prevent PsychologicalWar.org from treating “propaganda,” “psychological operations,” “information operations,” “strategic communication,” “public diplomacy,” “political advertising,” “public-information campaigns,” “behavioral public policy,” “misinformation,” and “disinformation” as interchangeable.

REQUIRED RESEARCH QUESTIONS

1. How do international organizations, regional bodies, governments, militaries, regulators, courts, scholars, journalists, and civil-society organizations define the principal terms?
2. Which activities are overt public communication, which are covert or attributed influence, which are ordinary diplomacy or education, and which are regulated political speech or advertising?
3. How do peacetime, armed-conflict, emergency, election, public-health, and commercial contexts change the applicable legal analysis?
4. Where do freedom of expression, privacy, data protection, consumer protection, political participation, cultural rights, and national-security law intersect?
5. How do different legal traditions handle attribution, labeling, state-media status, foreign-agent rules, political advertising, government speech, and counter-disinformation programs?
6. Which popular definitions are historically specific, translated poorly, or used strategically by interested actors?
7. What public-facing terminology can explain the field accurately without normalizing manipulation or becoming an operational guide?

MINIMUM COMPARATIVE COVERAGE

Use at least 18 legal or institutional systems across all inhabited regions. Include at minimum: two African systems, two West Asian or North African systems, two South Asian systems, two East or Southeast Asian systems, two European systems, two Latin American or Caribbean systems, two Pacific systems, one North American system, two small states or territories, and three multinational or regional organizations. Include both democratic and non-democratic systems without using those labels as complete explanations.

ASSIGNMENT-SPECIFIC DELIVERABLES

In addition to the universal output contract, produce:

- `global-terminology-crosswalk.csv` with term, native/local term, institution using it, legal or doctrinal source, scope, attribution status, rights implications, and translation cautions.
- `legal-context-matrix.csv` covering peacetime, armed conflict, elections, public health, emergencies, commercial persuasion, and education.
- `terminology-decision-tree.json` that helps an editor choose a precise descriptive term without making legal conclusions.
- A publication plan for 10–14 pages, including a central glossary, legal boundaries, myths, and comparative examples.
- A short “words to avoid or qualify” editorial register.

QUALITY GATE

Do not define the field from a single military doctrine or English-language tradition. Distinguish descriptive terminology from legal status. Do not imply that every persuasive message is propaganda or that every state communication is covert. Do not turn doctrine into tactics.

FINAL INSTRUCTION

Return the complete packet under the working filename `01-global-psychological-influence-vocabulary-and-law.md`. End with one concise next-research action that follows from the evidence gaps you found. Do not claim publication approval, legal approval, or editorial acceptance.

---

Prompt 2: Comparative Media Ecosystems, Ownership, Regulation, and Public-Service Communication

You are an independent international research agent. You have no access to the PsychologicalWar.org source code, repository, private files, admin panel, analytics, editorial memory, prior agent reports, or other agents. Do not request those materials and do not assume another agent will complete missing work. Research this assignment from public sources and return a self-contained, publication-ready research packet that a separate editorial team can use without contacting you.

ACCESS AND INDEPENDENCE

- Work from scratch using lawful, publicly accessible sources.
- Do not claim that you reviewed the website or its code.
- Do not rely on hidden project context, proprietary datasets, or unpublished instructions.
- Do not communicate with other research agents.
- Do not ask clarifying questions. Resolve ambiguity conservatively, state assumptions, and document unresolved questions.
- Return content, evidence, and editorial recommendations only. Do not write or request software code.

COMMISSIONING CONTEXT

PsychologicalWar.org is an independent international educational publication about psychological influence, information environments, public reasoning, institutional behavior, resilience, and the human consequences of information conflict. It is not a government agency, intelligence service, military organization, university, or the United States Intelligence Advanced Research Projects Activity. Do not imply sponsorship, endorsement, access, authority, or affiliation with any government or institution.

The publication already has introductory material on international intelligence institutions, world context, crisis resilience, formal versus practical authority, domestic politics of alignment, small-state agency, strategic dependencies, intelligence oversight, simulation literacy, and responsible virtual-society design. Your assignment must expand the public knowledge base without depending on that material and without merely restating it.

MANDATORY INTERNATIONAL FAIRNESS METHOD

1. Use equal analytical method, not artificial equivalence. Apply the same questions and evidence rules to every jurisdiction while preserving genuine differences.
2. Institutions are not populations. Never treat a government, service, ruling party, military, company, platform, religion, ethnicity, language group, nationality, migrant community, diaspora, or political movement as the inherent character of an entire population.
3. Separate formal authority from practical influence; law from implementation; policy from capacity; ownership from control; access from action; capability from intent; and institutional claims from independent findings.
4. Separate observation, source report, allegation, assumption, inference, hypothesis, assessment, decision, and confirmed outcome.
5. Separate probability from evidentiary confidence.
6. Treat uncertainty, evidence gaps, contestation, and source asymmetry as information. Do not fill gaps with stereotypes, neighboring-country practice, regional averages, or major-power analogies.
7. Do not interpret a larger public record of failures as proof of worse performance. Courts, audits, disclosure laws, journalism, and public archives can make transparent systems look more failure-prone than opaque systems.
8. Preserve local terminology, native scripts, name order, transliteration, and legally significant distinctions. Explain when a familiar English term is misleading.
9. Include small states, territories, associated states, multilingual societies, low-bandwidth environments, and nontraditional institutions with the same structural dignity given to major powers.
10. Keep citizenship, residence, birthplace, territorial identity, language, ethnicity, religion, migration, political opinion, and diaspora ties analytically separate.
11. Include rights, oversight, remedy, correction, appeal, and accountability where relevant. Formal safeguards must not be presented as proof of practical accessibility or effectiveness.
12. Do not create country rankings, civilization scales, morality scores, threat scores, trustworthiness scores, sophistication scores, development-by-nature claims, or winners and losers.

CURRENTNESS AND FRESHNESS

- Use the actual completion date as the research cutoff and state it in UTC.
- Reverify current laws, officeholders, regulators, institutional names, court decisions, platform policies, treaties, and active programs near the end of the research process.
- Place volatile facts in a clearly marked Current-Status Audit.
- Tag major fields as relatively stable, law-sensitive, officeholder-sensitive, policy-sensitive, platform-sensitive, conflict-sensitive, data-sensitive, or low-visibility.
- Record the last verification date for each current-sensitive claim.
- Distinguish event date, publication date, effective date, commencement date, and access date.
- When current verification is unavailable, state “Unknown” rather than guessing.

SOURCE AND CITATION RULES

Prioritize sources in this order:

1. Constitutions, statutes, regulations, official gazettes, court judgments, parliamentary records, regulator decisions, public audits, ombuds reports, official statistics, official institutional publications, and formal policy documents.
2. United Nations bodies, regional organizations, international courts, treaty bodies, standards organizations, and documented intergovernmental records.
3. Peer-reviewed scholarship, university presses, locally authored research, and historically grounded monographs.
4. High-quality journalism with transparent sourcing and visible corrections practices.
5. Reputable human-rights, civil-liberties, press-freedom, consumer-protection, accessibility, anti-corruption, and community organizations for documented concerns.
6. Technically rigorous platform, cybersecurity, data-governance, and media-forensics research for technical claims, while separating technical indicators from conclusions about human or state intent.

Additional source requirements:

- Use local-language or locally authoritative sources for every case whenever reasonably accessible.
- At least one primary or locally authoritative source must support each major case study.
- Trace apparently independent reports to their originating evidence so repetition is not mistaken for corroboration.
- Do not cite search-result snippets as evidence.
- Do not use Wikipedia as an authority; it may only help locate stronger sources.
- Attribute disputed claims to the actor making them.
- Provide inline citations close to the claims they support.
- Include a source register listing title, publisher, author when available, language, publication date, event date when different, URL, access date, source tier, jurisdiction, claim supported, and independence notes.
- Keep quotations brief and necessary. Prefer accurate paraphrase.

CLAIM LABELS

Apply one of these labels to every high-impact, contested, or current-sensitive claim and explain the basis:

- Officially confirmed
- Confirmed by multiple independent sources
- Strongly assessed
- Plausible but not conclusively demonstrated
- Credibly alleged
- Disputed
- Low confidence
- Unknown
- Outdated or superseded

SAFETY AND NON-OPERATIONAL BOUNDARY

This is public education, not an influence-operations manual. Do not provide:

- Targeting criteria for real individuals, communities, institutions, or facilities
- Step-by-step persuasion, coercion, radicalization, recruitment, deception, harassment, surveillance, infiltration, sabotage, evasion, concealment, or psychological manipulation procedures
- Advice for exploiting trauma, mental illness, religion, ethnicity, poverty, disability, migration status, sexuality, age, or political identity
- Microtargeting recipes, vulnerability scoring, covert persona construction, astroturfing procedures, bot deployment instructions, platform-evasion methods, or disinformation campaign design
- Private personal data, live travel patterns, residential details, or protective-security vulnerabilities
- Cyber intrusion, credential theft, malware, exploit chains, or methods to defeat moderation, law enforcement, border, or safety controls
- Diagnostic claims about individuals or populations
- Material that could reasonably enable real-world harm

You may explain institutions, historical events, public communication systems, research findings, safeguards, failure modes, and non-operational educational abstractions.

UNIVERSAL OUTPUT CONTRACT

Return one self-contained Markdown packet. Use the exact working filename specified in the assignment. Aim for 8,000–12,000 words excluding appendices and the source register. Include:

1. Title, status, exact UTC research cutoff, and a one-paragraph independence statement
2. Executive summary
3. Research questions, scope, exclusions, and definitions
4. Methodology, source hierarchy, geographic-selection logic, and confidence framework
5. Current-Status Audit
6. Substantive comparative analysis
7. Cross-regional case studies
8. Rights, accountability, remedy, correction, and accessibility analysis
9. Evidence asymmetry and source-limitations section
10. Common myths and evidence-based corrections
11. Research gaps and unresolved questions
12. Freshness and correction register
13. Publication plan containing proposed public page titles, route slugs, 120–180 word abstracts, section outlines, and source needs
14. Twelve direct-answer FAQs suitable for a general audience
15. Glossary of at least 30 terms
16. Complete source register
17. Machine-readable appendices as fenced CSV and JSON blocks

Do not include website code. Where multiple artifacts are requested, place them inside the same response under clearly labeled headings and give each a filename.

ASSIGNMENT

Research how information reaches the public through public-service media, state media, commercial outlets, community media, independent journalism, religious or customary communication channels, messaging applications, and platform-based distribution. Explain how funding, ownership, regulation, language, geography, infrastructure, and political practice shape media systems without assigning a single media-freedom or trust score.

REQUIRED RESEARCH QUESTIONS

1. How are public broadcasters, state broadcasters, commercial networks, community stations, and nonprofit outlets legally constituted and financed?
2. How do ownership concentration, cross-ownership, advertising dependence, public subsidy, licensing, and distribution infrastructure affect editorial independence?
3. What is the difference between formal regulator independence and practical political or commercial influence?
4. How do community radio, local-language media, public notice systems, and customary institutions serve populations underserved by national outlets?
5. How do defamation law, secrecy law, licensing, emergency rules, platform policy, and physical safety affect journalism in practice?
6. How do audiences combine television, radio, print, websites, messaging apps, diaspora media, and interpersonal networks?
7. What remedies exist for correction, reply, complaint, access, accessibility, and discriminatory exclusion?

MINIMUM COMPARATIVE COVERAGE

Select 20 jurisdictions using geographic and institutional diversity rather than media-market size. Include at least four small states or territories, four multilingual systems, three federal or highly decentralized systems, three low-bandwidth or island systems, two conflict-affected systems, and four jurisdictions with strong community-media traditions. Include at least six locally published or native-language source streams.

ASSIGNMENT-SPECIFIC DELIVERABLES

Also produce:

- `media-system-comparison.csv` with ownership, funding, regulator, public-service mandate, community-media role, dominant distribution channels, correction mechanisms, accessibility, and evidence limitations.
- `audience-pathways.json` describing non-personalized information routes such as broadcast, community, platform, messaging, and interpersonal channels.
- Twelve myth/correction pairs, including corrections to “state-funded means state-controlled,” “private means independent,” and “low internet penetration means low information circulation.”
- A publication plan for 12–16 pages, including regional examples and a page on evidence asymmetry.

QUALITY GATE

Do not treat media systems as national personalities. Avoid reducing entire countries to “free” or “controlled.” Separate legal structure, ownership, editorial practice, audience behavior, and individual journalist safety.

FINAL INSTRUCTION

Return the complete packet under the working filename `02-comparative-media-ecosystems-and-public-communication.md`. End with one concise next-research action that follows from the evidence gaps you found. Do not claim publication approval, legal approval, or editorial acceptance.

---

Prompt 3: Global Historical Atlas of Information Influence, Public Persuasion, and Institutional Correction

You are an independent international research agent. You have no access to the PsychologicalWar.org source code, repository, private files, admin panel, analytics, editorial memory, prior agent reports, or other agents. Do not request those materials and do not assume another agent will complete missing work. Research this assignment from public sources and return a self-contained, publication-ready research packet that a separate editorial team can use without contacting you.

ACCESS AND INDEPENDENCE

- Work from scratch using lawful, publicly accessible sources.
- Do not claim that you reviewed the website or its code.
- Do not rely on hidden project context, proprietary datasets, or unpublished instructions.
- Do not communicate with other research agents.
- Do not ask clarifying questions. Resolve ambiguity conservatively, state assumptions, and document unresolved questions.
- Return content, evidence, and editorial recommendations only. Do not write or request software code.

COMMISSIONING CONTEXT

PsychologicalWar.org is an independent international educational publication about psychological influence, information environments, public reasoning, institutional behavior, resilience, and the human consequences of information conflict. It is not a government agency, intelligence service, military organization, university, or the United States Intelligence Advanced Research Projects Activity. Do not imply sponsorship, endorsement, access, authority, or affiliation with any government or institution.

The publication already has introductory material on international intelligence institutions, world context, crisis resilience, formal versus practical authority, domestic politics of alignment, small-state agency, strategic dependencies, intelligence oversight, simulation literacy, and responsible virtual-society design. Your assignment must expand the public knowledge base without depending on that material and without merely restating it.

MANDATORY INTERNATIONAL FAIRNESS METHOD

1. Use equal analytical method, not artificial equivalence. Apply the same questions and evidence rules to every jurisdiction while preserving genuine differences.
2. Institutions are not populations. Never treat a government, service, ruling party, military, company, platform, religion, ethnicity, language group, nationality, migrant community, diaspora, or political movement as the inherent character of an entire population.
3. Separate formal authority from practical influence; law from implementation; policy from capacity; ownership from control; access from action; capability from intent; and institutional claims from independent findings.
4. Separate observation, source report, allegation, assumption, inference, hypothesis, assessment, decision, and confirmed outcome.
5. Separate probability from evidentiary confidence.
6. Treat uncertainty, evidence gaps, contestation, and source asymmetry as information. Do not fill gaps with stereotypes, neighboring-country practice, regional averages, or major-power analogies.
7. Do not interpret a larger public record of failures as proof of worse performance. Courts, audits, disclosure laws, journalism, and public archives can make transparent systems look more failure-prone than opaque systems.
8. Preserve local terminology, native scripts, name order, transliteration, and legally significant distinctions. Explain when a familiar English term is misleading.
9. Include small states, territories, associated states, multilingual societies, low-bandwidth environments, and nontraditional institutions with the same structural dignity given to major powers.
10. Keep citizenship, residence, birthplace, territorial identity, language, ethnicity, religion, migration, political opinion, and diaspora ties analytically separate.
11. Include rights, oversight, remedy, correction, appeal, and accountability where relevant. Formal safeguards must not be presented as proof of practical accessibility or effectiveness.
12. Do not create country rankings, civilization scales, morality scores, threat scores, trustworthiness scores, sophistication scores, development-by-nature claims, or winners and losers.

CURRENTNESS AND FRESHNESS

- Use the actual completion date as the research cutoff and state it in UTC.
- Reverify current laws, officeholders, regulators, institutional names, court decisions, platform policies, treaties, and active programs near the end of the research process.
- Place volatile facts in a clearly marked Current-Status Audit.
- Tag major fields as relatively stable, law-sensitive, officeholder-sensitive, policy-sensitive, platform-sensitive, conflict-sensitive, data-sensitive, or low-visibility.
- Record the last verification date for each current-sensitive claim.
- Distinguish event date, publication date, effective date, commencement date, and access date.
- When current verification is unavailable, state “Unknown” rather than guessing.

SOURCE AND CITATION RULES

Prioritize sources in this order:

1. Constitutions, statutes, regulations, official gazettes, court judgments, parliamentary records, regulator decisions, public audits, ombuds reports, official statistics, official institutional publications, and formal policy documents.
2. United Nations bodies, regional organizations, international courts, treaty bodies, standards organizations, and documented intergovernmental records.
3. Peer-reviewed scholarship, university presses, locally authored research, and historically grounded monographs.
4. High-quality journalism with transparent sourcing and visible corrections practices.
5. Reputable human-rights, civil-liberties, press-freedom, consumer-protection, accessibility, anti-corruption, and community organizations for documented concerns.
6. Technically rigorous platform, cybersecurity, data-governance, and media-forensics research for technical claims, while separating technical indicators from conclusions about human or state intent.

Additional source requirements:

- Use local-language or locally authoritative sources for every case whenever reasonably accessible.
- At least one primary or locally authoritative source must support each major case study.
- Trace apparently independent reports to their originating evidence so repetition is not mistaken for corroboration.
- Do not cite search-result snippets as evidence.
- Do not use Wikipedia as an authority; it may only help locate stronger sources.
- Attribute disputed claims to the actor making them.
- Provide inline citations close to the claims they support.
- Include a source register listing title, publisher, author when available, language, publication date, event date when different, URL, access date, source tier, jurisdiction, claim supported, and independence notes.
- Keep quotations brief and necessary. Prefer accurate paraphrase.

CLAIM LABELS

Apply one of these labels to every high-impact, contested, or current-sensitive claim and explain the basis:

- Officially confirmed
- Confirmed by multiple independent sources
- Strongly assessed
- Plausible but not conclusively demonstrated
- Credibly alleged
- Disputed
- Low confidence
- Unknown
- Outdated or superseded

SAFETY AND NON-OPERATIONAL BOUNDARY

This is public education, not an influence-operations manual. Do not provide:

- Targeting criteria for real individuals, communities, institutions, or facilities
- Step-by-step persuasion, coercion, radicalization, recruitment, deception, harassment, surveillance, infiltration, sabotage, evasion, concealment, or psychological manipulation procedures
- Advice for exploiting trauma, mental illness, religion, ethnicity, poverty, disability, migration status, sexuality, age, or political identity
- Microtargeting recipes, vulnerability scoring, covert persona construction, astroturfing procedures, bot deployment instructions, platform-evasion methods, or disinformation campaign design
- Private personal data, live travel patterns, residential details, or protective-security vulnerabilities
- Cyber intrusion, credential theft, malware, exploit chains, or methods to defeat moderation, law enforcement, border, or safety controls
- Diagnostic claims about individuals or populations
- Material that could reasonably enable real-world harm

You may explain institutions, historical events, public communication systems, research findings, safeguards, failure modes, and non-operational educational abstractions.

UNIVERSAL OUTPUT CONTRACT

Return one self-contained Markdown packet. Use the exact working filename specified in the assignment. Aim for 8,000–12,000 words excluding appendices and the source register. Include:

1. Title, status, exact UTC research cutoff, and a one-paragraph independence statement
2. Executive summary
3. Research questions, scope, exclusions, and definitions
4. Methodology, source hierarchy, geographic-selection logic, and confidence framework
5. Current-Status Audit
6. Substantive comparative analysis
7. Cross-regional case studies
8. Rights, accountability, remedy, correction, and accessibility analysis
9. Evidence asymmetry and source-limitations section
10. Common myths and evidence-based corrections
11. Research gaps and unresolved questions
12. Freshness and correction register
13. Publication plan containing proposed public page titles, route slugs, 120–180 word abstracts, section outlines, and source needs
14. Twelve direct-answer FAQs suitable for a general audience
15. Glossary of at least 30 terms
16. Complete source register
17. Machine-readable appendices as fenced CSV and JSON blocks

Do not include website code. Where multiple artifacts are requested, place them inside the same response under clearly labeled headings and give each a filename.

ASSIGNMENT

Build a balanced historical atlas of information influence, public persuasion, rumor, propaganda, strategic communication, and subsequent correction or accountability. The purpose is to teach readers how institutions, media systems, social conditions, and evidence quality shape outcomes—not to extract reusable manipulation tactics.

REQUIRED RESEARCH QUESTIONS

1. What was the documented institutional objective, audience context, communication environment, and historical setting of each case?
2. Which claims about the case are established, disputed, mythologized, or based on retrospective interpretation?
3. What role did censorship, journalism, literacy, language, technology, diaspora networks, public trust, and local intermediaries play?
4. How did people interpret, resist, adapt, parody, ignore, or repurpose official messaging?
5. What harms, unintended effects, policy failures, or rights violations followed?
6. What inquiries, archives, court cases, truth commissions, apologies, declassifications, corrections, or scholarly revisions changed the public record?
7. Which lessons concern institutional design and media literacy rather than operational technique?

MINIMUM COMPARATIVE COVERAGE

Include at least 30 cases spanning the nineteenth, twentieth, and twenty-first centuries. No region may supply more than 20 percent of the cases. Include at least six cases from Africa, five from Asia, four from West Asia or North Africa, five from Europe, five from Latin America or the Caribbean, three from Oceania or the Pacific, and two from North America. Include small-state, colonial, anticolonial, public-health, labor, disaster, peacebuilding, and commercial cases—not only wars and major powers. At least one third of the cases must emphasize correction, resistance, public learning, or institutional reform.

ASSIGNMENT-SPECIFIC DELIVERABLES

Also produce:

- `historical-case-register.csv` with case, dates, region, actors, audience context, medium, objective, evidence status, harms, correction mechanism, and present-day interpretation cautions.
- `myth-versus-record.csv` tracing popular stories to primary or scholarly evidence.
- `timeline.json` separating event date, publication date, disclosure date, and correction date.
- A publication plan for a 20-page case-study atlas plus five thematic synthesis pages.

QUALITY GATE

Do not romanticize “successful” influence, reproduce slogans at length, or provide a recipe for replication. Include the agency of audiences and affected communities. Treat historical government archives as evidence of institutional perspective, not automatic truth.

FINAL INSTRUCTION

Return the complete packet under the working filename `03-global-historical-information-influence-case-atlas.md`. End with one concise next-research action that follows from the evidence gaps you found. Do not claim publication approval, legal approval, or editorial acceptance.

---

Prompt 4: Rumor, Misinformation, Disinformation, and Crisis Communication Across Societies

You are an independent international research agent. You have no access to the PsychologicalWar.org source code, repository, private files, admin panel, analytics, editorial memory, prior agent reports, or other agents. Do not request those materials and do not assume another agent will complete missing work. Research this assignment from public sources and return a self-contained, publication-ready research packet that a separate editorial team can use without contacting you.

ACCESS AND INDEPENDENCE

- Work from scratch using lawful, publicly accessible sources.
- Do not claim that you reviewed the website or its code.
- Do not rely on hidden project context, proprietary datasets, or unpublished instructions.
- Do not communicate with other research agents.
- Do not ask clarifying questions. Resolve ambiguity conservatively, state assumptions, and document unresolved questions.
- Return content, evidence, and editorial recommendations only. Do not write or request software code.

COMMISSIONING CONTEXT

PsychologicalWar.org is an independent international educational publication about psychological influence, information environments, public reasoning, institutional behavior, resilience, and the human consequences of information conflict. It is not a government agency, intelligence service, military organization, university, or the United States Intelligence Advanced Research Projects Activity. Do not imply sponsorship, endorsement, access, authority, or affiliation with any government or institution.

The publication already has introductory material on international intelligence institutions, world context, crisis resilience, formal versus practical authority, domestic politics of alignment, small-state agency, strategic dependencies, intelligence oversight, simulation literacy, and responsible virtual-society design. Your assignment must expand the public knowledge base without depending on that material and without merely restating it.

MANDATORY INTERNATIONAL FAIRNESS METHOD

1. Use equal analytical method, not artificial equivalence. Apply the same questions and evidence rules to every jurisdiction while preserving genuine differences.
2. Institutions are not populations. Never treat a government, service, ruling party, military, company, platform, religion, ethnicity, language group, nationality, migrant community, diaspora, or political movement as the inherent character of an entire population.
3. Separate formal authority from practical influence; law from implementation; policy from capacity; ownership from control; access from action; capability from intent; and institutional claims from independent findings.
4. Separate observation, source report, allegation, assumption, inference, hypothesis, assessment, decision, and confirmed outcome.
5. Separate probability from evidentiary confidence.
6. Treat uncertainty, evidence gaps, contestation, and source asymmetry as information. Do not fill gaps with stereotypes, neighboring-country practice, regional averages, or major-power analogies.
7. Do not interpret a larger public record of failures as proof of worse performance. Courts, audits, disclosure laws, journalism, and public archives can make transparent systems look more failure-prone than opaque systems.
8. Preserve local terminology, native scripts, name order, transliteration, and legally significant distinctions. Explain when a familiar English term is misleading.
9. Include small states, territories, associated states, multilingual societies, low-bandwidth environments, and nontraditional institutions with the same structural dignity given to major powers.
10. Keep citizenship, residence, birthplace, territorial identity, language, ethnicity, religion, migration, political opinion, and diaspora ties analytically separate.
11. Include rights, oversight, remedy, correction, appeal, and accountability where relevant. Formal safeguards must not be presented as proof of practical accessibility or effectiveness.
12. Do not create country rankings, civilization scales, morality scores, threat scores, trustworthiness scores, sophistication scores, development-by-nature claims, or winners and losers.

CURRENTNESS AND FRESHNESS

- Use the actual completion date as the research cutoff and state it in UTC.
- Reverify current laws, officeholders, regulators, institutional names, court decisions, platform policies, treaties, and active programs near the end of the research process.
- Place volatile facts in a clearly marked Current-Status Audit.
- Tag major fields as relatively stable, law-sensitive, officeholder-sensitive, policy-sensitive, platform-sensitive, conflict-sensitive, data-sensitive, or low-visibility.
- Record the last verification date for each current-sensitive claim.
- Distinguish event date, publication date, effective date, commencement date, and access date.
- When current verification is unavailable, state “Unknown” rather than guessing.

SOURCE AND CITATION RULES

Prioritize sources in this order:

1. Constitutions, statutes, regulations, official gazettes, court judgments, parliamentary records, regulator decisions, public audits, ombuds reports, official statistics, official institutional publications, and formal policy documents.
2. United Nations bodies, regional organizations, international courts, treaty bodies, standards organizations, and documented intergovernmental records.
3. Peer-reviewed scholarship, university presses, locally authored research, and historically grounded monographs.
4. High-quality journalism with transparent sourcing and visible corrections practices.
5. Reputable human-rights, civil-liberties, press-freedom, consumer-protection, accessibility, anti-corruption, and community organizations for documented concerns.
6. Technically rigorous platform, cybersecurity, data-governance, and media-forensics research for technical claims, while separating technical indicators from conclusions about human or state intent.

Additional source requirements:

- Use local-language or locally authoritative sources for every case whenever reasonably accessible.
- At least one primary or locally authoritative source must support each major case study.
- Trace apparently independent reports to their originating evidence so repetition is not mistaken for corroboration.
- Do not cite search-result snippets as evidence.
- Do not use Wikipedia as an authority; it may only help locate stronger sources.
- Attribute disputed claims to the actor making them.
- Provide inline citations close to the claims they support.
- Include a source register listing title, publisher, author when available, language, publication date, event date when different, URL, access date, source tier, jurisdiction, claim supported, and independence notes.
- Keep quotations brief and necessary. Prefer accurate paraphrase.

CLAIM LABELS

Apply one of these labels to every high-impact, contested, or current-sensitive claim and explain the basis:

- Officially confirmed
- Confirmed by multiple independent sources
- Strongly assessed
- Plausible but not conclusively demonstrated
- Credibly alleged
- Disputed
- Low confidence
- Unknown
- Outdated or superseded

SAFETY AND NON-OPERATIONAL BOUNDARY

This is public education, not an influence-operations manual. Do not provide:

- Targeting criteria for real individuals, communities, institutions, or facilities
- Step-by-step persuasion, coercion, radicalization, recruitment, deception, harassment, surveillance, infiltration, sabotage, evasion, concealment, or psychological manipulation procedures
- Advice for exploiting trauma, mental illness, religion, ethnicity, poverty, disability, migration status, sexuality, age, or political identity
- Microtargeting recipes, vulnerability scoring, covert persona construction, astroturfing procedures, bot deployment instructions, platform-evasion methods, or disinformation campaign design
- Private personal data, live travel patterns, residential details, or protective-security vulnerabilities
- Cyber intrusion, credential theft, malware, exploit chains, or methods to defeat moderation, law enforcement, border, or safety controls
- Diagnostic claims about individuals or populations
- Material that could reasonably enable real-world harm

You may explain institutions, historical events, public communication systems, research findings, safeguards, failure modes, and non-operational educational abstractions.

UNIVERSAL OUTPUT CONTRACT

Return one self-contained Markdown packet. Use the exact working filename specified in the assignment. Aim for 8,000–12,000 words excluding appendices and the source register. Include:

1. Title, status, exact UTC research cutoff, and a one-paragraph independence statement
2. Executive summary
3. Research questions, scope, exclusions, and definitions
4. Methodology, source hierarchy, geographic-selection logic, and confidence framework
5. Current-Status Audit
6. Substantive comparative analysis
7. Cross-regional case studies
8. Rights, accountability, remedy, correction, and accessibility analysis
9. Evidence asymmetry and source-limitations section
10. Common myths and evidence-based corrections
11. Research gaps and unresolved questions
12. Freshness and correction register
13. Publication plan containing proposed public page titles, route slugs, 120–180 word abstracts, section outlines, and source needs
14. Twelve direct-answer FAQs suitable for a general audience
15. Glossary of at least 30 terms
16. Complete source register
17. Machine-readable appendices as fenced CSV and JSON blocks

Do not include website code. Where multiple artifacts are requested, place them inside the same response under clearly labeled headings and give each a filename.

ASSIGNMENT

Research how uncertain, false, misleading, incomplete, or rapidly changing information moves during disasters, epidemics, conflicts, communal tension, economic shocks, infrastructure failures, and other crises. Explain the difference between rumor, error, satire, misunderstanding, deliberate deception, evolving evidence, and contested interpretation.

REQUIRED RESEARCH QUESTIONS

1. How do information voids, fear, prior experience, institutional credibility, language barriers, and platform design affect rumor circulation?
2. How can authorities communicate uncertainty without creating false reassurance or panic?
3. What makes correction timely, trusted, accessible, and locally meaningful?
4. When do centralized messages fail because local conditions differ?
5. What roles do health workers, emergency services, journalists, community leaders, religious institutions, mutual-aid groups, and diaspora networks play?
6. How do internet shutdowns, power outages, damaged infrastructure, inaccessible formats, and low literacy alter crisis communication?
7. How should institutions document errors, update guidance, preserve prior versions, and compensate for harmful misinformation?

MINIMUM COMPARATIVE COVERAGE

Use at least 20 crisis cases across all regions, including public health, storms or floods, wildfire, earthquake, industrial accident, food or water disruption, conflict displacement, financial panic, and communication outage. Include at least six small-state, island, rural, Indigenous, or low-bandwidth contexts. Include cases where the official account was wrong or changed, not only cases of public misinformation.

ASSIGNMENT-SPECIFIC DELIVERABLES

Also produce:

- `crisis-information-case-register.csv`
- `message-correction-lifecycle.json` covering notice, uncertainty, update, withdrawal, archive, remedy, and review
- A non-operational public checklist for evaluating crisis claims
- A publication plan for 12–15 pages, including rumor, uncertainty, correction, accessibility, and community-ground-truth pages
- Twelve example rewrites showing how to replace false certainty with accurate uncertainty language

QUALITY GATE

Do not provide techniques for manufacturing panic, exploiting emergencies, evading moderation, or targeting vulnerable groups. Do not assume official information is correct or that community information is unreliable; evaluate each claim by evidence.

FINAL INSTRUCTION

Return the complete packet under the working filename `04-rumor-misinformation-and-crisis-communication.md`. End with one concise next-research action that follows from the evidence gaps you found. Do not claim publication approval, legal approval, or editorial acceptance.

---

Prompt 5: Language, Translation, Symbolism, and Cross-Cultural Meaning in Public Communication

You are an independent international research agent. You have no access to the PsychologicalWar.org source code, repository, private files, admin panel, analytics, editorial memory, prior agent reports, or other agents. Do not request those materials and do not assume another agent will complete missing work. Research this assignment from public sources and return a self-contained, publication-ready research packet that a separate editorial team can use without contacting you.

ACCESS AND INDEPENDENCE

- Work from scratch using lawful, publicly accessible sources.
- Do not claim that you reviewed the website or its code.
- Do not rely on hidden project context, proprietary datasets, or unpublished instructions.
- Do not communicate with other research agents.
- Do not ask clarifying questions. Resolve ambiguity conservatively, state assumptions, and document unresolved questions.
- Return content, evidence, and editorial recommendations only. Do not write or request software code.

COMMISSIONING CONTEXT

PsychologicalWar.org is an independent international educational publication about psychological influence, information environments, public reasoning, institutional behavior, resilience, and the human consequences of information conflict. It is not a government agency, intelligence service, military organization, university, or the United States Intelligence Advanced Research Projects Activity. Do not imply sponsorship, endorsement, access, authority, or affiliation with any government or institution.

The publication already has introductory material on international intelligence institutions, world context, crisis resilience, formal versus practical authority, domestic politics of alignment, small-state agency, strategic dependencies, intelligence oversight, simulation literacy, and responsible virtual-society design. Your assignment must expand the public knowledge base without depending on that material and without merely restating it.

MANDATORY INTERNATIONAL FAIRNESS METHOD

1. Use equal analytical method, not artificial equivalence. Apply the same questions and evidence rules to every jurisdiction while preserving genuine differences.
2. Institutions are not populations. Never treat a government, service, ruling party, military, company, platform, religion, ethnicity, language group, nationality, migrant community, diaspora, or political movement as the inherent character of an entire population.
3. Separate formal authority from practical influence; law from implementation; policy from capacity; ownership from control; access from action; capability from intent; and institutional claims from independent findings.
4. Separate observation, source report, allegation, assumption, inference, hypothesis, assessment, decision, and confirmed outcome.
5. Separate probability from evidentiary confidence.
6. Treat uncertainty, evidence gaps, contestation, and source asymmetry as information. Do not fill gaps with stereotypes, neighboring-country practice, regional averages, or major-power analogies.
7. Do not interpret a larger public record of failures as proof of worse performance. Courts, audits, disclosure laws, journalism, and public archives can make transparent systems look more failure-prone than opaque systems.
8. Preserve local terminology, native scripts, name order, transliteration, and legally significant distinctions. Explain when a familiar English term is misleading.
9. Include small states, territories, associated states, multilingual societies, low-bandwidth environments, and nontraditional institutions with the same structural dignity given to major powers.
10. Keep citizenship, residence, birthplace, territorial identity, language, ethnicity, religion, migration, political opinion, and diaspora ties analytically separate.
11. Include rights, oversight, remedy, correction, appeal, and accountability where relevant. Formal safeguards must not be presented as proof of practical accessibility or effectiveness.
12. Do not create country rankings, civilization scales, morality scores, threat scores, trustworthiness scores, sophistication scores, development-by-nature claims, or winners and losers.

CURRENTNESS AND FRESHNESS

- Use the actual completion date as the research cutoff and state it in UTC.
- Reverify current laws, officeholders, regulators, institutional names, court decisions, platform policies, treaties, and active programs near the end of the research process.
- Place volatile facts in a clearly marked Current-Status Audit.
- Tag major fields as relatively stable, law-sensitive, officeholder-sensitive, policy-sensitive, platform-sensitive, conflict-sensitive, data-sensitive, or low-visibility.
- Record the last verification date for each current-sensitive claim.
- Distinguish event date, publication date, effective date, commencement date, and access date.
- When current verification is unavailable, state “Unknown” rather than guessing.

SOURCE AND CITATION RULES

Prioritize sources in this order:

1. Constitutions, statutes, regulations, official gazettes, court judgments, parliamentary records, regulator decisions, public audits, ombuds reports, official statistics, official institutional publications, and formal policy documents.
2. United Nations bodies, regional organizations, international courts, treaty bodies, standards organizations, and documented intergovernmental records.
3. Peer-reviewed scholarship, university presses, locally authored research, and historically grounded monographs.
4. High-quality journalism with transparent sourcing and visible corrections practices.
5. Reputable human-rights, civil-liberties, press-freedom, consumer-protection, accessibility, anti-corruption, and community organizations for documented concerns.
6. Technically rigorous platform, cybersecurity, data-governance, and media-forensics research for technical claims, while separating technical indicators from conclusions about human or state intent.

Additional source requirements:

- Use local-language or locally authoritative sources for every case whenever reasonably accessible.
- At least one primary or locally authoritative source must support each major case study.
- Trace apparently independent reports to their originating evidence so repetition is not mistaken for corroboration.
- Do not cite search-result snippets as evidence.
- Do not use Wikipedia as an authority; it may only help locate stronger sources.
- Attribute disputed claims to the actor making them.
- Provide inline citations close to the claims they support.
- Include a source register listing title, publisher, author when available, language, publication date, event date when different, URL, access date, source tier, jurisdiction, claim supported, and independence notes.
- Keep quotations brief and necessary. Prefer accurate paraphrase.

CLAIM LABELS

Apply one of these labels to every high-impact, contested, or current-sensitive claim and explain the basis:

- Officially confirmed
- Confirmed by multiple independent sources
- Strongly assessed
- Plausible but not conclusively demonstrated
- Credibly alleged
- Disputed
- Low confidence
- Unknown
- Outdated or superseded

SAFETY AND NON-OPERATIONAL BOUNDARY

This is public education, not an influence-operations manual. Do not provide:

- Targeting criteria for real individuals, communities, institutions, or facilities
- Step-by-step persuasion, coercion, radicalization, recruitment, deception, harassment, surveillance, infiltration, sabotage, evasion, concealment, or psychological manipulation procedures
- Advice for exploiting trauma, mental illness, religion, ethnicity, poverty, disability, migration status, sexuality, age, or political identity
- Microtargeting recipes, vulnerability scoring, covert persona construction, astroturfing procedures, bot deployment instructions, platform-evasion methods, or disinformation campaign design
- Private personal data, live travel patterns, residential details, or protective-security vulnerabilities
- Cyber intrusion, credential theft, malware, exploit chains, or methods to defeat moderation, law enforcement, border, or safety controls
- Diagnostic claims about individuals or populations
- Material that could reasonably enable real-world harm

You may explain institutions, historical events, public communication systems, research findings, safeguards, failure modes, and non-operational educational abstractions.

UNIVERSAL OUTPUT CONTRACT

Return one self-contained Markdown packet. Use the exact working filename specified in the assignment. Aim for 8,000–12,000 words excluding appendices and the source register. Include:

1. Title, status, exact UTC research cutoff, and a one-paragraph independence statement
2. Executive summary
3. Research questions, scope, exclusions, and definitions
4. Methodology, source hierarchy, geographic-selection logic, and confidence framework
5. Current-Status Audit
6. Substantive comparative analysis
7. Cross-regional case studies
8. Rights, accountability, remedy, correction, and accessibility analysis
9. Evidence asymmetry and source-limitations section
10. Common myths and evidence-based corrections
11. Research gaps and unresolved questions
12. Freshness and correction register
13. Publication plan containing proposed public page titles, route slugs, 120–180 word abstracts, section outlines, and source needs
14. Twelve direct-answer FAQs suitable for a general audience
15. Glossary of at least 30 terms
16. Complete source register
17. Machine-readable appendices as fenced CSV and JSON blocks

Do not include website code. Where multiple artifacts are requested, place them inside the same response under clearly labeled headings and give each a filename.

ASSIGNMENT

Research how meaning changes across languages, scripts, dialects, registers, translation systems, visual symbols, humor, metaphor, gesture, and cultural context. The goal is to improve internationally respectful public content and prevent translation errors from being mistaken for intent, hostility, irrationality, or deception.

REQUIRED RESEARCH QUESTIONS

1. How do literal translation, legal translation, journalistic translation, simultaneous interpretation, and machine translation differ?
2. How do honorifics, pronouns, evidential markers, politeness, indirectness, code-switching, diglossia, and name order affect interpretation?
3. What is lost when native scripts, diacritics, right-to-left layout, or culturally specific terms are normalized into English-only fields?
4. How do symbols, colors, maps, historical references, humor, irony, and gesture change across contexts?
5. How can editors distinguish a translation disagreement from factual contradiction or deceptive intent?
6. What review practices reduce error when no single translation is definitive?
7. How should public pages display original wording, transliteration, translation, uncertainty, and competing interpretations accessibly?

MINIMUM COMPARATIVE COVERAGE

Use at least 24 linguistic contexts across at least 12 language families and all inhabited regions. Include right-to-left scripts, abugidas, syllabaries, logographic writing, creoles, sign languages, Indigenous languages, endangered languages, and multilingual administrative systems. Include at least eight real translation or interpretation disputes documented by courts, diplomacy, journalism, public health, or emergency communication.

ASSIGNMENT-SPECIFIC DELIVERABLES

Also produce:

- `translation-risk-register.csv`
- `script-and-name-display-guide.csv`
- `translation-evidence-model.json`
- A publication plan for 12–16 pages
- An editorial checklist for native script, transliteration, translation provenance, reviewer qualifications, bidirectional text, pronunciation, and accessibility
- Twelve annotated examples showing how apparently equivalent English words diverge in legal or institutional meaning

QUALITY GATE

Avoid cultural personality claims such as “direct cultures” or “secretive cultures.” Explain communication practices as context-dependent, internally diverse, and shaped by institutions, relationships, and situation.

FINAL INSTRUCTION

Return the complete packet under the working filename `05-language-translation-and-cross-cultural-meaning.md`. End with one concise next-research action that follows from the evidence gaps you found. Do not claim publication approval, legal approval, or editorial acceptance.

---

Prompt 6: Public Opinion, Polling, Surveys, and the Limits of Measuring Belief

You are an independent international research agent. You have no access to the PsychologicalWar.org source code, repository, private files, admin panel, analytics, editorial memory, prior agent reports, or other agents. Do not request those materials and do not assume another agent will complete missing work. Research this assignment from public sources and return a self-contained, publication-ready research packet that a separate editorial team can use without contacting you.

ACCESS AND INDEPENDENCE

- Work from scratch using lawful, publicly accessible sources.
- Do not claim that you reviewed the website or its code.
- Do not rely on hidden project context, proprietary datasets, or unpublished instructions.
- Do not communicate with other research agents.
- Do not ask clarifying questions. Resolve ambiguity conservatively, state assumptions, and document unresolved questions.
- Return content, evidence, and editorial recommendations only. Do not write or request software code.

COMMISSIONING CONTEXT

PsychologicalWar.org is an independent international educational publication about psychological influence, information environments, public reasoning, institutional behavior, resilience, and the human consequences of information conflict. It is not a government agency, intelligence service, military organization, university, or the United States Intelligence Advanced Research Projects Activity. Do not imply sponsorship, endorsement, access, authority, or affiliation with any government or institution.

The publication already has introductory material on international intelligence institutions, world context, crisis resilience, formal versus practical authority, domestic politics of alignment, small-state agency, strategic dependencies, intelligence oversight, simulation literacy, and responsible virtual-society design. Your assignment must expand the public knowledge base without depending on that material and without merely restating it.

MANDATORY INTERNATIONAL FAIRNESS METHOD

1. Use equal analytical method, not artificial equivalence. Apply the same questions and evidence rules to every jurisdiction while preserving genuine differences.
2. Institutions are not populations. Never treat a government, service, ruling party, military, company, platform, religion, ethnicity, language group, nationality, migrant community, diaspora, or political movement as the inherent character of an entire population.
3. Separate formal authority from practical influence; law from implementation; policy from capacity; ownership from control; access from action; capability from intent; and institutional claims from independent findings.
4. Separate observation, source report, allegation, assumption, inference, hypothesis, assessment, decision, and confirmed outcome.
5. Separate probability from evidentiary confidence.
6. Treat uncertainty, evidence gaps, contestation, and source asymmetry as information. Do not fill gaps with stereotypes, neighboring-country practice, regional averages, or major-power analogies.
7. Do not interpret a larger public record of failures as proof of worse performance. Courts, audits, disclosure laws, journalism, and public archives can make transparent systems look more failure-prone than opaque systems.
8. Preserve local terminology, native scripts, name order, transliteration, and legally significant distinctions. Explain when a familiar English term is misleading.
9. Include small states, territories, associated states, multilingual societies, low-bandwidth environments, and nontraditional institutions with the same structural dignity given to major powers.
10. Keep citizenship, residence, birthplace, territorial identity, language, ethnicity, religion, migration, political opinion, and diaspora ties analytically separate.
11. Include rights, oversight, remedy, correction, appeal, and accountability where relevant. Formal safeguards must not be presented as proof of practical accessibility or effectiveness.
12. Do not create country rankings, civilization scales, morality scores, threat scores, trustworthiness scores, sophistication scores, development-by-nature claims, or winners and losers.

CURRENTNESS AND FRESHNESS

- Use the actual completion date as the research cutoff and state it in UTC.
- Reverify current laws, officeholders, regulators, institutional names, court decisions, platform policies, treaties, and active programs near the end of the research process.
- Place volatile facts in a clearly marked Current-Status Audit.
- Tag major fields as relatively stable, law-sensitive, officeholder-sensitive, policy-sensitive, platform-sensitive, conflict-sensitive, data-sensitive, or low-visibility.
- Record the last verification date for each current-sensitive claim.
- Distinguish event date, publication date, effective date, commencement date, and access date.
- When current verification is unavailable, state “Unknown” rather than guessing.

SOURCE AND CITATION RULES

Prioritize sources in this order:

1. Constitutions, statutes, regulations, official gazettes, court judgments, parliamentary records, regulator decisions, public audits, ombuds reports, official statistics, official institutional publications, and formal policy documents.
2. United Nations bodies, regional organizations, international courts, treaty bodies, standards organizations, and documented intergovernmental records.
3. Peer-reviewed scholarship, university presses, locally authored research, and historically grounded monographs.
4. High-quality journalism with transparent sourcing and visible corrections practices.
5. Reputable human-rights, civil-liberties, press-freedom, consumer-protection, accessibility, anti-corruption, and community organizations for documented concerns.
6. Technically rigorous platform, cybersecurity, data-governance, and media-forensics research for technical claims, while separating technical indicators from conclusions about human or state intent.

Additional source requirements:

- Use local-language or locally authoritative sources for every case whenever reasonably accessible.
- At least one primary or locally authoritative source must support each major case study.
- Trace apparently independent reports to their originating evidence so repetition is not mistaken for corroboration.
- Do not cite search-result snippets as evidence.
- Do not use Wikipedia as an authority; it may only help locate stronger sources.
- Attribute disputed claims to the actor making them.
- Provide inline citations close to the claims they support.
- Include a source register listing title, publisher, author when available, language, publication date, event date when different, URL, access date, source tier, jurisdiction, claim supported, and independence notes.
- Keep quotations brief and necessary. Prefer accurate paraphrase.

CLAIM LABELS

Apply one of these labels to every high-impact, contested, or current-sensitive claim and explain the basis:

- Officially confirmed
- Confirmed by multiple independent sources
- Strongly assessed
- Plausible but not conclusively demonstrated
- Credibly alleged
- Disputed
- Low confidence
- Unknown
- Outdated or superseded

SAFETY AND NON-OPERATIONAL BOUNDARY

This is public education, not an influence-operations manual. Do not provide:

- Targeting criteria for real individuals, communities, institutions, or facilities
- Step-by-step persuasion, coercion, radicalization, recruitment, deception, harassment, surveillance, infiltration, sabotage, evasion, concealment, or psychological manipulation procedures
- Advice for exploiting trauma, mental illness, religion, ethnicity, poverty, disability, migration status, sexuality, age, or political identity
- Microtargeting recipes, vulnerability scoring, covert persona construction, astroturfing procedures, bot deployment instructions, platform-evasion methods, or disinformation campaign design
- Private personal data, live travel patterns, residential details, or protective-security vulnerabilities
- Cyber intrusion, credential theft, malware, exploit chains, or methods to defeat moderation, law enforcement, border, or safety controls
- Diagnostic claims about individuals or populations
- Material that could reasonably enable real-world harm

You may explain institutions, historical events, public communication systems, research findings, safeguards, failure modes, and non-operational educational abstractions.

UNIVERSAL OUTPUT CONTRACT

Return one self-contained Markdown packet. Use the exact working filename specified in the assignment. Aim for 8,000–12,000 words excluding appendices and the source register. Include:

1. Title, status, exact UTC research cutoff, and a one-paragraph independence statement
2. Executive summary
3. Research questions, scope, exclusions, and definitions
4. Methodology, source hierarchy, geographic-selection logic, and confidence framework
5. Current-Status Audit
6. Substantive comparative analysis
7. Cross-regional case studies
8. Rights, accountability, remedy, correction, and accessibility analysis
9. Evidence asymmetry and source-limitations section
10. Common myths and evidence-based corrections
11. Research gaps and unresolved questions
12. Freshness and correction register
13. Publication plan containing proposed public page titles, route slugs, 120–180 word abstracts, section outlines, and source needs
14. Twelve direct-answer FAQs suitable for a general audience
15. Glossary of at least 30 terms
16. Complete source register
17. Machine-readable appendices as fenced CSV and JSON blocks

Do not include website code. Where multiple artifacts are requested, place them inside the same response under clearly labeled headings and give each a filename.

ASSIGNMENT

Research how public opinion is measured, reported, misunderstood, and sometimes weaponized. Build a globally useful guide to polling, surveys, focus groups, administrative data, social-media signals, and qualitative research that prevents readers from treating a single percentage as the voice of a whole country.

REQUIRED RESEARCH QUESTIONS

1. How do sampling frames, nonresponse, weighting, question order, wording, translation, mode, timing, and sponsor affect results?
2. How do fear, censorship, social desirability, conflict, displacement, low connectivity, and distrust change who answers and what they say?
3. What are the limits of online opt-in polls, platform sentiment analysis, search trends, and social-media data?
4. How should uncertainty intervals, design effects, subgroup sample sizes, missing data, and repeated measures be communicated?
5. How do national surveys handle multilingual, Indigenous, nomadic, migrant, refugee, stateless, remote, and institutionalized populations?
6. How can editors separate public opinion from government policy, election outcome, media visibility, and diaspora advocacy?
7. What correction duties arise when a poll is misreported or its fieldwork becomes obsolete?

MINIMUM COMPARATIVE COVERAGE

Compare at least 18 survey environments across all regions. Include democratic, authoritarian, conflict-affected, federal, multilingual, small-island, low-connectivity, refugee-hosting, and territorially contested contexts. Include at least six examples where headline reporting overstated what the underlying survey supported.

ASSIGNMENT-SPECIFIC DELIVERABLES

Also produce:

- `poll-quality-checklist.csv`
- `survey-claim-register.json`
- Twelve worked examples that reconstruct a headline from the underlying methodology
- A publication plan for 10–14 pages
- A plain-language guide to margins of error, uncertainty intervals, weighting, nonresponse, subgroup limits, and longitudinal comparison

QUALITY GATE

Do not infer fixed national psychology from survey results. Do not use polling to classify populations as loyal, hostile, irrational, or susceptible. Distinguish measured response, unmeasured population, interpretation, and policy consequence.

FINAL INSTRUCTION

Return the complete packet under the working filename `06-public-opinion-polling-and-measurement-literacy.md`. End with one concise next-research action that follows from the evidence gaps you found. Do not claim publication approval, legal approval, or editorial acceptance.

---

Prompt 7: Digital Platforms, Recommender Systems, Synthetic Media, and AI-Mediated Influence

You are an independent international research agent. You have no access to the PsychologicalWar.org source code, repository, private files, admin panel, analytics, editorial memory, prior agent reports, or other agents. Do not request those materials and do not assume another agent will complete missing work. Research this assignment from public sources and return a self-contained, publication-ready research packet that a separate editorial team can use without contacting you.

ACCESS AND INDEPENDENCE

- Work from scratch using lawful, publicly accessible sources.
- Do not claim that you reviewed the website or its code.
- Do not rely on hidden project context, proprietary datasets, or unpublished instructions.
- Do not communicate with other research agents.
- Do not ask clarifying questions. Resolve ambiguity conservatively, state assumptions, and document unresolved questions.
- Return content, evidence, and editorial recommendations only. Do not write or request software code.

COMMISSIONING CONTEXT

PsychologicalWar.org is an independent international educational publication about psychological influence, information environments, public reasoning, institutional behavior, resilience, and the human consequences of information conflict. It is not a government agency, intelligence service, military organization, university, or the United States Intelligence Advanced Research Projects Activity. Do not imply sponsorship, endorsement, access, authority, or affiliation with any government or institution.

The publication already has introductory material on international intelligence institutions, world context, crisis resilience, formal versus practical authority, domestic politics of alignment, small-state agency, strategic dependencies, intelligence oversight, simulation literacy, and responsible virtual-society design. Your assignment must expand the public knowledge base without depending on that material and without merely restating it.

MANDATORY INTERNATIONAL FAIRNESS METHOD

1. Use equal analytical method, not artificial equivalence. Apply the same questions and evidence rules to every jurisdiction while preserving genuine differences.
2. Institutions are not populations. Never treat a government, service, ruling party, military, company, platform, religion, ethnicity, language group, nationality, migrant community, diaspora, or political movement as the inherent character of an entire population.
3. Separate formal authority from practical influence; law from implementation; policy from capacity; ownership from control; access from action; capability from intent; and institutional claims from independent findings.
4. Separate observation, source report, allegation, assumption, inference, hypothesis, assessment, decision, and confirmed outcome.
5. Separate probability from evidentiary confidence.
6. Treat uncertainty, evidence gaps, contestation, and source asymmetry as information. Do not fill gaps with stereotypes, neighboring-country practice, regional averages, or major-power analogies.
7. Do not interpret a larger public record of failures as proof of worse performance. Courts, audits, disclosure laws, journalism, and public archives can make transparent systems look more failure-prone than opaque systems.
8. Preserve local terminology, native scripts, name order, transliteration, and legally significant distinctions. Explain when a familiar English term is misleading.
9. Include small states, territories, associated states, multilingual societies, low-bandwidth environments, and nontraditional institutions with the same structural dignity given to major powers.
10. Keep citizenship, residence, birthplace, territorial identity, language, ethnicity, religion, migration, political opinion, and diaspora ties analytically separate.
11. Include rights, oversight, remedy, correction, appeal, and accountability where relevant. Formal safeguards must not be presented as proof of practical accessibility or effectiveness.
12. Do not create country rankings, civilization scales, morality scores, threat scores, trustworthiness scores, sophistication scores, development-by-nature claims, or winners and losers.

CURRENTNESS AND FRESHNESS

- Use the actual completion date as the research cutoff and state it in UTC.
- Reverify current laws, officeholders, regulators, institutional names, court decisions, platform policies, treaties, and active programs near the end of the research process.
- Place volatile facts in a clearly marked Current-Status Audit.
- Tag major fields as relatively stable, law-sensitive, officeholder-sensitive, policy-sensitive, platform-sensitive, conflict-sensitive, data-sensitive, or low-visibility.
- Record the last verification date for each current-sensitive claim.
- Distinguish event date, publication date, effective date, commencement date, and access date.
- When current verification is unavailable, state “Unknown” rather than guessing.

SOURCE AND CITATION RULES

Prioritize sources in this order:

1. Constitutions, statutes, regulations, official gazettes, court judgments, parliamentary records, regulator decisions, public audits, ombuds reports, official statistics, official institutional publications, and formal policy documents.
2. United Nations bodies, regional organizations, international courts, treaty bodies, standards organizations, and documented intergovernmental records.
3. Peer-reviewed scholarship, university presses, locally authored research, and historically grounded monographs.
4. High-quality journalism with transparent sourcing and visible corrections practices.
5. Reputable human-rights, civil-liberties, press-freedom, consumer-protection, accessibility, anti-corruption, and community organizations for documented concerns.
6. Technically rigorous platform, cybersecurity, data-governance, and media-forensics research for technical claims, while separating technical indicators from conclusions about human or state intent.

Additional source requirements:

- Use local-language or locally authoritative sources for every case whenever reasonably accessible.
- At least one primary or locally authoritative source must support each major case study.
- Trace apparently independent reports to their originating evidence so repetition is not mistaken for corroboration.
- Do not cite search-result snippets as evidence.
- Do not use Wikipedia as an authority; it may only help locate stronger sources.
- Attribute disputed claims to the actor making them.
- Provide inline citations close to the claims they support.
- Include a source register listing title, publisher, author when available, language, publication date, event date when different, URL, access date, source tier, jurisdiction, claim supported, and independence notes.
- Keep quotations brief and necessary. Prefer accurate paraphrase.

CLAIM LABELS

Apply one of these labels to every high-impact, contested, or current-sensitive claim and explain the basis:

- Officially confirmed
- Confirmed by multiple independent sources
- Strongly assessed
- Plausible but not conclusively demonstrated
- Credibly alleged
- Disputed
- Low confidence
- Unknown
- Outdated or superseded

SAFETY AND NON-OPERATIONAL BOUNDARY

This is public education, not an influence-operations manual. Do not provide:

- Targeting criteria for real individuals, communities, institutions, or facilities
- Step-by-step persuasion, coercion, radicalization, recruitment, deception, harassment, surveillance, infiltration, sabotage, evasion, concealment, or psychological manipulation procedures
- Advice for exploiting trauma, mental illness, religion, ethnicity, poverty, disability, migration status, sexuality, age, or political identity
- Microtargeting recipes, vulnerability scoring, covert persona construction, astroturfing procedures, bot deployment instructions, platform-evasion methods, or disinformation campaign design
- Private personal data, live travel patterns, residential details, or protective-security vulnerabilities
- Cyber intrusion, credential theft, malware, exploit chains, or methods to defeat moderation, law enforcement, border, or safety controls
- Diagnostic claims about individuals or populations
- Material that could reasonably enable real-world harm

You may explain institutions, historical events, public communication systems, research findings, safeguards, failure modes, and non-operational educational abstractions.

UNIVERSAL OUTPUT CONTRACT

Return one self-contained Markdown packet. Use the exact working filename specified in the assignment. Aim for 8,000–12,000 words excluding appendices and the source register. Include:

1. Title, status, exact UTC research cutoff, and a one-paragraph independence statement
2. Executive summary
3. Research questions, scope, exclusions, and definitions
4. Methodology, source hierarchy, geographic-selection logic, and confidence framework
5. Current-Status Audit
6. Substantive comparative analysis
7. Cross-regional case studies
8. Rights, accountability, remedy, correction, and accessibility analysis
9. Evidence asymmetry and source-limitations section
10. Common myths and evidence-based corrections
11. Research gaps and unresolved questions
12. Freshness and correction register
13. Publication plan containing proposed public page titles, route slugs, 120–180 word abstracts, section outlines, and source needs
14. Twelve direct-answer FAQs suitable for a general audience
15. Glossary of at least 30 terms
16. Complete source register
17. Machine-readable appendices as fenced CSV and JSON blocks

Do not include website code. Where multiple artifacts are requested, place them inside the same response under clearly labeled headings and give each a filename.

ASSIGNMENT

Research how social platforms, search engines, messaging services, recommendation systems, generative AI, synthetic audio and video, advertising exchanges, and content-moderation systems shape what people encounter. Focus on governance, incentives, transparency, provenance, user rights, and public understanding—not on techniques for manipulating platforms.

REQUIRED RESEARCH QUESTIONS

1. How do ranking, recommendation, virality, advertising, creator monetization, forwarding, and closed-group messaging differ as distribution systems?
2. What can researchers legitimately infer from engagement metrics, and what remains unknown?
3. How do synthetic media, voice cloning, automated translation, AI summaries, and generated search results alter provenance and correction?
4. What labeling, watermarking, authenticity, content-credential, and disclosure approaches exist, and what are their limitations?
5. How do privacy, age assurance, accessibility, consumer protection, political advertising, competition law, and freedom of expression interact?
6. How do platform rules and enforcement differ across languages and regions?
7. What appeal, explanation, data-access, correction, and independent-audit mechanisms are available to users and researchers?

MINIMUM COMPARATIVE COVERAGE

Compare at least 12 regulatory or legal systems, six platform or service types, and 18 documented cases from all regions. Include low-resource-language moderation, small-market regulation, encrypted messaging, public broadcasters using platforms, youth protection, disability access, and cross-border enforcement. Verify current rules and effective dates.

ASSIGNMENT-SPECIFIC DELIVERABLES

Also produce:

- `platform-governance-comparison.csv`
- `synthetic-media-provenance-model.json`
- `user-rights-and-remedies.csv`
- A publication plan for 14–18 pages
- A non-technical explainer of recommendation systems that avoids implying a single secret “algorithm”
- Twelve myths and corrections, including “engagement equals persuasion,” “viral equals representative,” and “watermarked equals true”

QUALITY GATE

Do not provide botting, astroturfing, prompt-injection, moderation-evasion, deepfake production, microtargeting, or covert amplification procedures. Treat platform claims, researcher claims, and regulator findings as distinct evidence categories.

FINAL INSTRUCTION

Return the complete packet under the working filename `07-platforms-recommenders-synthetic-media-and-ai-influence.md`. End with one concise next-research action that follows from the evidence gaps you found. Do not claim publication approval, legal approval, or editorial acceptance.

---

Prompt 8: Journalism, Fact-Checking, Whistleblowing, Archives, and Public Correction Systems

You are an independent international research agent. You have no access to the PsychologicalWar.org source code, repository, private files, admin panel, analytics, editorial memory, prior agent reports, or other agents. Do not request those materials and do not assume another agent will complete missing work. Research this assignment from public sources and return a self-contained, publication-ready research packet that a separate editorial team can use without contacting you.

ACCESS AND INDEPENDENCE

- Work from scratch using lawful, publicly accessible sources.
- Do not claim that you reviewed the website or its code.
- Do not rely on hidden project context, proprietary datasets, or unpublished instructions.
- Do not communicate with other research agents.
- Do not ask clarifying questions. Resolve ambiguity conservatively, state assumptions, and document unresolved questions.
- Return content, evidence, and editorial recommendations only. Do not write or request software code.

COMMISSIONING CONTEXT

PsychologicalWar.org is an independent international educational publication about psychological influence, information environments, public reasoning, institutional behavior, resilience, and the human consequences of information conflict. It is not a government agency, intelligence service, military organization, university, or the United States Intelligence Advanced Research Projects Activity. Do not imply sponsorship, endorsement, access, authority, or affiliation with any government or institution.

The publication already has introductory material on international intelligence institutions, world context, crisis resilience, formal versus practical authority, domestic politics of alignment, small-state agency, strategic dependencies, intelligence oversight, simulation literacy, and responsible virtual-society design. Your assignment must expand the public knowledge base without depending on that material and without merely restating it.

MANDATORY INTERNATIONAL FAIRNESS METHOD

1. Use equal analytical method, not artificial equivalence. Apply the same questions and evidence rules to every jurisdiction while preserving genuine differences.
2. Institutions are not populations. Never treat a government, service, ruling party, military, company, platform, religion, ethnicity, language group, nationality, migrant community, diaspora, or political movement as the inherent character of an entire population.
3. Separate formal authority from practical influence; law from implementation; policy from capacity; ownership from control; access from action; capability from intent; and institutional claims from independent findings.
4. Separate observation, source report, allegation, assumption, inference, hypothesis, assessment, decision, and confirmed outcome.
5. Separate probability from evidentiary confidence.
6. Treat uncertainty, evidence gaps, contestation, and source asymmetry as information. Do not fill gaps with stereotypes, neighboring-country practice, regional averages, or major-power analogies.
7. Do not interpret a larger public record of failures as proof of worse performance. Courts, audits, disclosure laws, journalism, and public archives can make transparent systems look more failure-prone than opaque systems.
8. Preserve local terminology, native scripts, name order, transliteration, and legally significant distinctions. Explain when a familiar English term is misleading.
9. Include small states, territories, associated states, multilingual societies, low-bandwidth environments, and nontraditional institutions with the same structural dignity given to major powers.
10. Keep citizenship, residence, birthplace, territorial identity, language, ethnicity, religion, migration, political opinion, and diaspora ties analytically separate.
11. Include rights, oversight, remedy, correction, appeal, and accountability where relevant. Formal safeguards must not be presented as proof of practical accessibility or effectiveness.
12. Do not create country rankings, civilization scales, morality scores, threat scores, trustworthiness scores, sophistication scores, development-by-nature claims, or winners and losers.

CURRENTNESS AND FRESHNESS

- Use the actual completion date as the research cutoff and state it in UTC.
- Reverify current laws, officeholders, regulators, institutional names, court decisions, platform policies, treaties, and active programs near the end of the research process.
- Place volatile facts in a clearly marked Current-Status Audit.
- Tag major fields as relatively stable, law-sensitive, officeholder-sensitive, policy-sensitive, platform-sensitive, conflict-sensitive, data-sensitive, or low-visibility.
- Record the last verification date for each current-sensitive claim.
- Distinguish event date, publication date, effective date, commencement date, and access date.
- When current verification is unavailable, state “Unknown” rather than guessing.

SOURCE AND CITATION RULES

Prioritize sources in this order:

1. Constitutions, statutes, regulations, official gazettes, court judgments, parliamentary records, regulator decisions, public audits, ombuds reports, official statistics, official institutional publications, and formal policy documents.
2. United Nations bodies, regional organizations, international courts, treaty bodies, standards organizations, and documented intergovernmental records.
3. Peer-reviewed scholarship, university presses, locally authored research, and historically grounded monographs.
4. High-quality journalism with transparent sourcing and visible corrections practices.
5. Reputable human-rights, civil-liberties, press-freedom, consumer-protection, accessibility, anti-corruption, and community organizations for documented concerns.
6. Technically rigorous platform, cybersecurity, data-governance, and media-forensics research for technical claims, while separating technical indicators from conclusions about human or state intent.

Additional source requirements:

- Use local-language or locally authoritative sources for every case whenever reasonably accessible.
- At least one primary or locally authoritative source must support each major case study.
- Trace apparently independent reports to their originating evidence so repetition is not mistaken for corroboration.
- Do not cite search-result snippets as evidence.
- Do not use Wikipedia as an authority; it may only help locate stronger sources.
- Attribute disputed claims to the actor making them.
- Provide inline citations close to the claims they support.
- Include a source register listing title, publisher, author when available, language, publication date, event date when different, URL, access date, source tier, jurisdiction, claim supported, and independence notes.
- Keep quotations brief and necessary. Prefer accurate paraphrase.

CLAIM LABELS

Apply one of these labels to every high-impact, contested, or current-sensitive claim and explain the basis:

- Officially confirmed
- Confirmed by multiple independent sources
- Strongly assessed
- Plausible but not conclusively demonstrated
- Credibly alleged
- Disputed
- Low confidence
- Unknown
- Outdated or superseded

SAFETY AND NON-OPERATIONAL BOUNDARY

This is public education, not an influence-operations manual. Do not provide:

- Targeting criteria for real individuals, communities, institutions, or facilities
- Step-by-step persuasion, coercion, radicalization, recruitment, deception, harassment, surveillance, infiltration, sabotage, evasion, concealment, or psychological manipulation procedures
- Advice for exploiting trauma, mental illness, religion, ethnicity, poverty, disability, migration status, sexuality, age, or political identity
- Microtargeting recipes, vulnerability scoring, covert persona construction, astroturfing procedures, bot deployment instructions, platform-evasion methods, or disinformation campaign design
- Private personal data, live travel patterns, residential details, or protective-security vulnerabilities
- Cyber intrusion, credential theft, malware, exploit chains, or methods to defeat moderation, law enforcement, border, or safety controls
- Diagnostic claims about individuals or populations
- Material that could reasonably enable real-world harm

You may explain institutions, historical events, public communication systems, research findings, safeguards, failure modes, and non-operational educational abstractions.

UNIVERSAL OUTPUT CONTRACT

Return one self-contained Markdown packet. Use the exact working filename specified in the assignment. Aim for 8,000–12,000 words excluding appendices and the source register. Include:

1. Title, status, exact UTC research cutoff, and a one-paragraph independence statement
2. Executive summary
3. Research questions, scope, exclusions, and definitions
4. Methodology, source hierarchy, geographic-selection logic, and confidence framework
5. Current-Status Audit
6. Substantive comparative analysis
7. Cross-regional case studies
8. Rights, accountability, remedy, correction, and accessibility analysis
9. Evidence asymmetry and source-limitations section
10. Common myths and evidence-based corrections
11. Research gaps and unresolved questions
12. Freshness and correction register
13. Publication plan containing proposed public page titles, route slugs, 120–180 word abstracts, section outlines, and source needs
14. Twelve direct-answer FAQs suitable for a general audience
15. Glossary of at least 30 terms
16. Complete source register
17. Machine-readable appendices as fenced CSV and JSON blocks

Do not include website code. Where multiple artifacts are requested, place them inside the same response under clearly labeled headings and give each a filename.

ASSIGNMENT

Research the institutions that help societies discover, contest, correct, preserve, and revisit public claims. Cover journalism, public records, freedom-of-information systems, fact-checking, press councils, ombuds institutions, whistleblower channels, archives, audit bodies, source protection, defamation law, anti-SLAPP mechanisms, and public-interest litigation.

REQUIRED RESEARCH QUESTIONS

1. How do correction policies, retractions, editor’s notes, right-of-reply rules, press councils, and broadcast regulators work?
2. What protects or exposes confidential sources and whistleblowers?
3. How do secrecy law, defamation, contempt, national-security exemptions, ownership pressure, violence, and economic precarity constrain reporting?
4. How do archives, declassification, court records, public audits, and inquiry reports change historical understanding?
5. What distinguishes fact-checking of a verifiable claim from adjudication of political values or unsettled science?
6. How do local, community, exile, diaspora, and cross-border investigative networks operate without treating any one form as inherently superior?
7. What remedies exist when reporting harms an individual or community, and what limitations remain?

MINIMUM COMPARATIVE COVERAGE

Compare at least 18 jurisdictions and six non-state or multinational accountability models. Include small states, territories, federal systems, multilingual systems, conflict-affected systems, and jurisdictions with limited public-record access. Include at least eight cases where a correction, inquiry, court ruling, archive release, or whistleblower disclosure materially changed the public record.

ASSIGNMENT-SPECIFIC DELIVERABLES

Also produce:

- `correction-and-accountability-mechanisms.csv`
- `source-protection-comparison.csv`
- `public-record-lifecycle.json`
- A publication plan for 12–16 pages
- Twelve case briefs showing the difference between allegation, publication, correction, adjudication, and historical reassessment

QUALITY GATE

Do not equate press volume with truth, state secrecy with competence, or a correction with institutional failure. Attribute findings precisely and distinguish protected disclosure from unlawful exposure of private people.

FINAL INSTRUCTION

Return the complete packet under the working filename `08-journalism-fact-checking-whistleblowing-and-correction.md`. End with one concise next-research action that follows from the evidence gaps you found. Do not claim publication approval, legal approval, or editorial acceptance.

---

Prompt 9: Diasporas, Migration, Remittances, and Transnational Information Networks

You are an independent international research agent. You have no access to the PsychologicalWar.org source code, repository, private files, admin panel, analytics, editorial memory, prior agent reports, or other agents. Do not request those materials and do not assume another agent will complete missing work. Research this assignment from public sources and return a self-contained, publication-ready research packet that a separate editorial team can use without contacting you.

ACCESS AND INDEPENDENCE

- Work from scratch using lawful, publicly accessible sources.
- Do not claim that you reviewed the website or its code.
- Do not rely on hidden project context, proprietary datasets, or unpublished instructions.
- Do not communicate with other research agents.
- Do not ask clarifying questions. Resolve ambiguity conservatively, state assumptions, and document unresolved questions.
- Return content, evidence, and editorial recommendations only. Do not write or request software code.

COMMISSIONING CONTEXT

PsychologicalWar.org is an independent international educational publication about psychological influence, information environments, public reasoning, institutional behavior, resilience, and the human consequences of information conflict. It is not a government agency, intelligence service, military organization, university, or the United States Intelligence Advanced Research Projects Activity. Do not imply sponsorship, endorsement, access, authority, or affiliation with any government or institution.

The publication already has introductory material on international intelligence institutions, world context, crisis resilience, formal versus practical authority, domestic politics of alignment, small-state agency, strategic dependencies, intelligence oversight, simulation literacy, and responsible virtual-society design. Your assignment must expand the public knowledge base without depending on that material and without merely restating it.

MANDATORY INTERNATIONAL FAIRNESS METHOD

1. Use equal analytical method, not artificial equivalence. Apply the same questions and evidence rules to every jurisdiction while preserving genuine differences.
2. Institutions are not populations. Never treat a government, service, ruling party, military, company, platform, religion, ethnicity, language group, nationality, migrant community, diaspora, or political movement as the inherent character of an entire population.
3. Separate formal authority from practical influence; law from implementation; policy from capacity; ownership from control; access from action; capability from intent; and institutional claims from independent findings.
4. Separate observation, source report, allegation, assumption, inference, hypothesis, assessment, decision, and confirmed outcome.
5. Separate probability from evidentiary confidence.
6. Treat uncertainty, evidence gaps, contestation, and source asymmetry as information. Do not fill gaps with stereotypes, neighboring-country practice, regional averages, or major-power analogies.
7. Do not interpret a larger public record of failures as proof of worse performance. Courts, audits, disclosure laws, journalism, and public archives can make transparent systems look more failure-prone than opaque systems.
8. Preserve local terminology, native scripts, name order, transliteration, and legally significant distinctions. Explain when a familiar English term is misleading.
9. Include small states, territories, associated states, multilingual societies, low-bandwidth environments, and nontraditional institutions with the same structural dignity given to major powers.
10. Keep citizenship, residence, birthplace, territorial identity, language, ethnicity, religion, migration, political opinion, and diaspora ties analytically separate.
11. Include rights, oversight, remedy, correction, appeal, and accountability where relevant. Formal safeguards must not be presented as proof of practical accessibility or effectiveness.
12. Do not create country rankings, civilization scales, morality scores, threat scores, trustworthiness scores, sophistication scores, development-by-nature claims, or winners and losers.

CURRENTNESS AND FRESHNESS

- Use the actual completion date as the research cutoff and state it in UTC.
- Reverify current laws, officeholders, regulators, institutional names, court decisions, platform policies, treaties, and active programs near the end of the research process.
- Place volatile facts in a clearly marked Current-Status Audit.
- Tag major fields as relatively stable, law-sensitive, officeholder-sensitive, policy-sensitive, platform-sensitive, conflict-sensitive, data-sensitive, or low-visibility.
- Record the last verification date for each current-sensitive claim.
- Distinguish event date, publication date, effective date, commencement date, and access date.
- When current verification is unavailable, state “Unknown” rather than guessing.

SOURCE AND CITATION RULES

Prioritize sources in this order:

1. Constitutions, statutes, regulations, official gazettes, court judgments, parliamentary records, regulator decisions, public audits, ombuds reports, official statistics, official institutional publications, and formal policy documents.
2. United Nations bodies, regional organizations, international courts, treaty bodies, standards organizations, and documented intergovernmental records.
3. Peer-reviewed scholarship, university presses, locally authored research, and historically grounded monographs.
4. High-quality journalism with transparent sourcing and visible corrections practices.
5. Reputable human-rights, civil-liberties, press-freedom, consumer-protection, accessibility, anti-corruption, and community organizations for documented concerns.
6. Technically rigorous platform, cybersecurity, data-governance, and media-forensics research for technical claims, while separating technical indicators from conclusions about human or state intent.

Additional source requirements:

- Use local-language or locally authoritative sources for every case whenever reasonably accessible.
- At least one primary or locally authoritative source must support each major case study.
- Trace apparently independent reports to their originating evidence so repetition is not mistaken for corroboration.
- Do not cite search-result snippets as evidence.
- Do not use Wikipedia as an authority; it may only help locate stronger sources.
- Attribute disputed claims to the actor making them.
- Provide inline citations close to the claims they support.
- Include a source register listing title, publisher, author when available, language, publication date, event date when different, URL, access date, source tier, jurisdiction, claim supported, and independence notes.
- Keep quotations brief and necessary. Prefer accurate paraphrase.

CLAIM LABELS

Apply one of these labels to every high-impact, contested, or current-sensitive claim and explain the basis:

- Officially confirmed
- Confirmed by multiple independent sources
- Strongly assessed
- Plausible but not conclusively demonstrated
- Credibly alleged
- Disputed
- Low confidence
- Unknown
- Outdated or superseded

SAFETY AND NON-OPERATIONAL BOUNDARY

This is public education, not an influence-operations manual. Do not provide:

- Targeting criteria for real individuals, communities, institutions, or facilities
- Step-by-step persuasion, coercion, radicalization, recruitment, deception, harassment, surveillance, infiltration, sabotage, evasion, concealment, or psychological manipulation procedures
- Advice for exploiting trauma, mental illness, religion, ethnicity, poverty, disability, migration status, sexuality, age, or political identity
- Microtargeting recipes, vulnerability scoring, covert persona construction, astroturfing procedures, bot deployment instructions, platform-evasion methods, or disinformation campaign design
- Private personal data, live travel patterns, residential details, or protective-security vulnerabilities
- Cyber intrusion, credential theft, malware, exploit chains, or methods to defeat moderation, law enforcement, border, or safety controls
- Diagnostic claims about individuals or populations
- Material that could reasonably enable real-world harm

You may explain institutions, historical events, public communication systems, research findings, safeguards, failure modes, and non-operational educational abstractions.

UNIVERSAL OUTPUT CONTRACT

Return one self-contained Markdown packet. Use the exact working filename specified in the assignment. Aim for 8,000–12,000 words excluding appendices and the source register. Include:

1. Title, status, exact UTC research cutoff, and a one-paragraph independence statement
2. Executive summary
3. Research questions, scope, exclusions, and definitions
4. Methodology, source hierarchy, geographic-selection logic, and confidence framework
5. Current-Status Audit
6. Substantive comparative analysis
7. Cross-regional case studies
8. Rights, accountability, remedy, correction, and accessibility analysis
9. Evidence asymmetry and source-limitations section
10. Common myths and evidence-based corrections
11. Research gaps and unresolved questions
12. Freshness and correction register
13. Publication plan containing proposed public page titles, route slugs, 120–180 word abstracts, section outlines, and source needs
14. Twelve direct-answer FAQs suitable for a general audience
15. Glossary of at least 30 terms
16. Complete source register
17. Machine-readable appendices as fenced CSV and JSON blocks

Do not include website code. Where multiple artifacts are requested, place them inside the same response under clearly labeled headings and give each a filename.

ASSIGNMENT

Research how people maintain information, family, economic, cultural, religious, political, and professional connections across borders. Explain diaspora and migrant information networks without treating communities as extensions of governments, intelligence services, political parties, or armed organizations.

REQUIRED RESEARCH QUESTIONS

1. How do family messaging, remittances, community associations, religious institutions, ethnic media, professional networks, student groups, and hometown organizations move information?
2. How do migrants and diasporas navigate multiple languages, media systems, legal statuses, and political contexts?
3. What is the evidentiary difference between cultural affiliation, political advocacy, lobbying, state outreach, coercion, surveillance allegations, and documented direction?
4. How do transnational repression claims, foreign-agent laws, consular services, dual citizenship, refugee protection, and asylum law intersect?
5. How do platform access, censorship, time zones, remittance corridors, and travel restrictions shape information flows?
6. How do diaspora outlets correct false claims and manage conflict imported from origin or host countries?
7. How can public content represent internal diversity, generational differences, mixed identities, and disagreement?

MINIMUM COMPARATIVE COVERAGE

Study at least 16 transnational corridors or communities across all regions. Include labor migration, refugee displacement, postcolonial ties, Indigenous cross-border communities, student migration, seafaring or mobile labor, and small-state diasporas. No corridor may be framed only through security concerns. Use sources produced by community organizations and scholars from the communities being discussed.

ASSIGNMENT-SPECIFIC DELIVERABLES

Also produce:

- `transnational-network-case-register.csv`
- `affiliation-direction-evidence-test.json`
- A publication plan for 12–15 pages
- Twelve myth/correction pairs addressing loyalty, dual citizenship, remittances, language retention, political disagreement, and state outreach
- An editorial checklist that prevents diaspora-as-proxy and migration-as-risk framing

QUALITY GATE

Cultural ties, family contact, language retention, travel, remittances, activism, or criticism are not evidence of state direction. Do not profile communities, infer loyalty, or describe populations as vectors, assets, fronts, or extensions of a state.

FINAL INSTRUCTION

Return the complete packet under the working filename `09-diasporas-migration-and-transnational-information-networks.md`. End with one concise next-research action that follows from the evidence gaps you found. Do not claim publication approval, legal approval, or editorial acceptance.

---

Prompt 10: Information Access, Censorship, Shutdowns, Digital Divides, and Accessibility

You are an independent international research agent. You have no access to the PsychologicalWar.org source code, repository, private files, admin panel, analytics, editorial memory, prior agent reports, or other agents. Do not request those materials and do not assume another agent will complete missing work. Research this assignment from public sources and return a self-contained, publication-ready research packet that a separate editorial team can use without contacting you.

ACCESS AND INDEPENDENCE

- Work from scratch using lawful, publicly accessible sources.
- Do not claim that you reviewed the website or its code.
- Do not rely on hidden project context, proprietary datasets, or unpublished instructions.
- Do not communicate with other research agents.
- Do not ask clarifying questions. Resolve ambiguity conservatively, state assumptions, and document unresolved questions.
- Return content, evidence, and editorial recommendations only. Do not write or request software code.

COMMISSIONING CONTEXT

PsychologicalWar.org is an independent international educational publication about psychological influence, information environments, public reasoning, institutional behavior, resilience, and the human consequences of information conflict. It is not a government agency, intelligence service, military organization, university, or the United States Intelligence Advanced Research Projects Activity. Do not imply sponsorship, endorsement, access, authority, or affiliation with any government or institution.

The publication already has introductory material on international intelligence institutions, world context, crisis resilience, formal versus practical authority, domestic politics of alignment, small-state agency, strategic dependencies, intelligence oversight, simulation literacy, and responsible virtual-society design. Your assignment must expand the public knowledge base without depending on that material and without merely restating it.

MANDATORY INTERNATIONAL FAIRNESS METHOD

1. Use equal analytical method, not artificial equivalence. Apply the same questions and evidence rules to every jurisdiction while preserving genuine differences.
2. Institutions are not populations. Never treat a government, service, ruling party, military, company, platform, religion, ethnicity, language group, nationality, migrant community, diaspora, or political movement as the inherent character of an entire population.
3. Separate formal authority from practical influence; law from implementation; policy from capacity; ownership from control; access from action; capability from intent; and institutional claims from independent findings.
4. Separate observation, source report, allegation, assumption, inference, hypothesis, assessment, decision, and confirmed outcome.
5. Separate probability from evidentiary confidence.
6. Treat uncertainty, evidence gaps, contestation, and source asymmetry as information. Do not fill gaps with stereotypes, neighboring-country practice, regional averages, or major-power analogies.
7. Do not interpret a larger public record of failures as proof of worse performance. Courts, audits, disclosure laws, journalism, and public archives can make transparent systems look more failure-prone than opaque systems.
8. Preserve local terminology, native scripts, name order, transliteration, and legally significant distinctions. Explain when a familiar English term is misleading.
9. Include small states, territories, associated states, multilingual societies, low-bandwidth environments, and nontraditional institutions with the same structural dignity given to major powers.
10. Keep citizenship, residence, birthplace, territorial identity, language, ethnicity, religion, migration, political opinion, and diaspora ties analytically separate.
11. Include rights, oversight, remedy, correction, appeal, and accountability where relevant. Formal safeguards must not be presented as proof of practical accessibility or effectiveness.
12. Do not create country rankings, civilization scales, morality scores, threat scores, trustworthiness scores, sophistication scores, development-by-nature claims, or winners and losers.

CURRENTNESS AND FRESHNESS

- Use the actual completion date as the research cutoff and state it in UTC.
- Reverify current laws, officeholders, regulators, institutional names, court decisions, platform policies, treaties, and active programs near the end of the research process.
- Place volatile facts in a clearly marked Current-Status Audit.
- Tag major fields as relatively stable, law-sensitive, officeholder-sensitive, policy-sensitive, platform-sensitive, conflict-sensitive, data-sensitive, or low-visibility.
- Record the last verification date for each current-sensitive claim.
- Distinguish event date, publication date, effective date, commencement date, and access date.
- When current verification is unavailable, state “Unknown” rather than guessing.

SOURCE AND CITATION RULES

Prioritize sources in this order:

1. Constitutions, statutes, regulations, official gazettes, court judgments, parliamentary records, regulator decisions, public audits, ombuds reports, official statistics, official institutional publications, and formal policy documents.
2. United Nations bodies, regional organizations, international courts, treaty bodies, standards organizations, and documented intergovernmental records.
3. Peer-reviewed scholarship, university presses, locally authored research, and historically grounded monographs.
4. High-quality journalism with transparent sourcing and visible corrections practices.
5. Reputable human-rights, civil-liberties, press-freedom, consumer-protection, accessibility, anti-corruption, and community organizations for documented concerns.
6. Technically rigorous platform, cybersecurity, data-governance, and media-forensics research for technical claims, while separating technical indicators from conclusions about human or state intent.

Additional source requirements:

- Use local-language or locally authoritative sources for every case whenever reasonably accessible.
- At least one primary or locally authoritative source must support each major case study.
- Trace apparently independent reports to their originating evidence so repetition is not mistaken for corroboration.
- Do not cite search-result snippets as evidence.
- Do not use Wikipedia as an authority; it may only help locate stronger sources.
- Attribute disputed claims to the actor making them.
- Provide inline citations close to the claims they support.
- Include a source register listing title, publisher, author when available, language, publication date, event date when different, URL, access date, source tier, jurisdiction, claim supported, and independence notes.
- Keep quotations brief and necessary. Prefer accurate paraphrase.

CLAIM LABELS

Apply one of these labels to every high-impact, contested, or current-sensitive claim and explain the basis:

- Officially confirmed
- Confirmed by multiple independent sources
- Strongly assessed
- Plausible but not conclusively demonstrated
- Credibly alleged
- Disputed
- Low confidence
- Unknown
- Outdated or superseded

SAFETY AND NON-OPERATIONAL BOUNDARY

This is public education, not an influence-operations manual. Do not provide:

- Targeting criteria for real individuals, communities, institutions, or facilities
- Step-by-step persuasion, coercion, radicalization, recruitment, deception, harassment, surveillance, infiltration, sabotage, evasion, concealment, or psychological manipulation procedures
- Advice for exploiting trauma, mental illness, religion, ethnicity, poverty, disability, migration status, sexuality, age, or political identity
- Microtargeting recipes, vulnerability scoring, covert persona construction, astroturfing procedures, bot deployment instructions, platform-evasion methods, or disinformation campaign design
- Private personal data, live travel patterns, residential details, or protective-security vulnerabilities
- Cyber intrusion, credential theft, malware, exploit chains, or methods to defeat moderation, law enforcement, border, or safety controls
- Diagnostic claims about individuals or populations
- Material that could reasonably enable real-world harm

You may explain institutions, historical events, public communication systems, research findings, safeguards, failure modes, and non-operational educational abstractions.

UNIVERSAL OUTPUT CONTRACT

Return one self-contained Markdown packet. Use the exact working filename specified in the assignment. Aim for 8,000–12,000 words excluding appendices and the source register. Include:

1. Title, status, exact UTC research cutoff, and a one-paragraph independence statement
2. Executive summary
3. Research questions, scope, exclusions, and definitions
4. Methodology, source hierarchy, geographic-selection logic, and confidence framework
5. Current-Status Audit
6. Substantive comparative analysis
7. Cross-regional case studies
8. Rights, accountability, remedy, correction, and accessibility analysis
9. Evidence asymmetry and source-limitations section
10. Common myths and evidence-based corrections
11. Research gaps and unresolved questions
12. Freshness and correction register
13. Publication plan containing proposed public page titles, route slugs, 120–180 word abstracts, section outlines, and source needs
14. Twelve direct-answer FAQs suitable for a general audience
15. Glossary of at least 30 terms
16. Complete source register
17. Machine-readable appendices as fenced CSV and JSON blocks

Do not include website code. Where multiple artifacts are requested, place them inside the same response under clearly labeled headings and give each a filename.

ASSIGNMENT

Research the material conditions that determine whether people can receive, verify, create, and correct information: electricity, network infrastructure, device access, cost, literacy, disability access, broadcasting, postal systems, libraries, community centers, language support, censorship, filtering, shutdowns, and emergency communications.

REQUIRED RESEARCH QUESTIONS

1. How do electricity reliability, mobile coverage, broadband, undersea cables, satellites, radio, print, and public institutions interact?
2. How do cost, data caps, device sharing, gendered access, age, disability, rurality, remoteness, displacement, and caregiving affect practical access?
3. What legal and administrative mechanisms authorize blocking, filtering, licensing, throttling, shutdowns, or emergency alerts?
4. How do formal accessibility standards compare with practical usability for screen readers, captions, sign language, low vision, cognitive accessibility, and low-bandwidth access?
5. What alternatives remain when internet or power fails?
6. How do people route around infrastructure failure through lawful community and humanitarian systems without turning the report into an evasion manual?
7. What complaint, judicial-review, regulator, or compensation mechanisms exist?

MINIMUM COMPARATIVE COVERAGE

Compare at least 20 jurisdictions, territories, or regions. Include landlocked, archipelagic, Arctic, desert, conflict-affected, high-income, low-income, Indigenous, and multilingual contexts. Include at least five internet or communications shutdown cases, five disaster or infrastructure-failure cases, and five accessibility or affordability cases.

ASSIGNMENT-SPECIFIC DELIVERABLES

Also produce:

- `information-access-comparison.csv`
- `shutdown-and-outage-case-register.csv`
- `accessible-publication-requirements.json`
- A publication plan for 14–18 pages
- A layered access model distinguishing legal availability, technical reach, affordability, language, accessibility, trust, and practical ability to act

QUALITY GATE

Do not treat connectivity as a development or civilization score. Do not publish sensitive network-node locations or bypass procedures. Separate ownership, operation, resilience, affordability, and user access.

FINAL INSTRUCTION

Return the complete packet under the working filename `10-information-access-censorship-shutdowns-and-accessibility.md`. End with one concise next-research action that follows from the evidence gaps you found. Do not claim publication approval, legal approval, or editorial acceptance.

---

Prompt 11: Civic Education, Media Literacy, Prebunking, Community Resilience, and Trust Repair

You are an independent international research agent. You have no access to the PsychologicalWar.org source code, repository, private files, admin panel, analytics, editorial memory, prior agent reports, or other agents. Do not request those materials and do not assume another agent will complete missing work. Research this assignment from public sources and return a self-contained, publication-ready research packet that a separate editorial team can use without contacting you.

ACCESS AND INDEPENDENCE

- Work from scratch using lawful, publicly accessible sources.
- Do not claim that you reviewed the website or its code.
- Do not rely on hidden project context, proprietary datasets, or unpublished instructions.
- Do not communicate with other research agents.
- Do not ask clarifying questions. Resolve ambiguity conservatively, state assumptions, and document unresolved questions.
- Return content, evidence, and editorial recommendations only. Do not write or request software code.

COMMISSIONING CONTEXT

PsychologicalWar.org is an independent international educational publication about psychological influence, information environments, public reasoning, institutional behavior, resilience, and the human consequences of information conflict. It is not a government agency, intelligence service, military organization, university, or the United States Intelligence Advanced Research Projects Activity. Do not imply sponsorship, endorsement, access, authority, or affiliation with any government or institution.

The publication already has introductory material on international intelligence institutions, world context, crisis resilience, formal versus practical authority, domestic politics of alignment, small-state agency, strategic dependencies, intelligence oversight, simulation literacy, and responsible virtual-society design. Your assignment must expand the public knowledge base without depending on that material and without merely restating it.

MANDATORY INTERNATIONAL FAIRNESS METHOD

1. Use equal analytical method, not artificial equivalence. Apply the same questions and evidence rules to every jurisdiction while preserving genuine differences.
2. Institutions are not populations. Never treat a government, service, ruling party, military, company, platform, religion, ethnicity, language group, nationality, migrant community, diaspora, or political movement as the inherent character of an entire population.
3. Separate formal authority from practical influence; law from implementation; policy from capacity; ownership from control; access from action; capability from intent; and institutional claims from independent findings.
4. Separate observation, source report, allegation, assumption, inference, hypothesis, assessment, decision, and confirmed outcome.
5. Separate probability from evidentiary confidence.
6. Treat uncertainty, evidence gaps, contestation, and source asymmetry as information. Do not fill gaps with stereotypes, neighboring-country practice, regional averages, or major-power analogies.
7. Do not interpret a larger public record of failures as proof of worse performance. Courts, audits, disclosure laws, journalism, and public archives can make transparent systems look more failure-prone than opaque systems.
8. Preserve local terminology, native scripts, name order, transliteration, and legally significant distinctions. Explain when a familiar English term is misleading.
9. Include small states, territories, associated states, multilingual societies, low-bandwidth environments, and nontraditional institutions with the same structural dignity given to major powers.
10. Keep citizenship, residence, birthplace, territorial identity, language, ethnicity, religion, migration, political opinion, and diaspora ties analytically separate.
11. Include rights, oversight, remedy, correction, appeal, and accountability where relevant. Formal safeguards must not be presented as proof of practical accessibility or effectiveness.
12. Do not create country rankings, civilization scales, morality scores, threat scores, trustworthiness scores, sophistication scores, development-by-nature claims, or winners and losers.

CURRENTNESS AND FRESHNESS

- Use the actual completion date as the research cutoff and state it in UTC.
- Reverify current laws, officeholders, regulators, institutional names, court decisions, platform policies, treaties, and active programs near the end of the research process.
- Place volatile facts in a clearly marked Current-Status Audit.
- Tag major fields as relatively stable, law-sensitive, officeholder-sensitive, policy-sensitive, platform-sensitive, conflict-sensitive, data-sensitive, or low-visibility.
- Record the last verification date for each current-sensitive claim.
- Distinguish event date, publication date, effective date, commencement date, and access date.
- When current verification is unavailable, state “Unknown” rather than guessing.

SOURCE AND CITATION RULES

Prioritize sources in this order:

1. Constitutions, statutes, regulations, official gazettes, court judgments, parliamentary records, regulator decisions, public audits, ombuds reports, official statistics, official institutional publications, and formal policy documents.
2. United Nations bodies, regional organizations, international courts, treaty bodies, standards organizations, and documented intergovernmental records.
3. Peer-reviewed scholarship, university presses, locally authored research, and historically grounded monographs.
4. High-quality journalism with transparent sourcing and visible corrections practices.
5. Reputable human-rights, civil-liberties, press-freedom, consumer-protection, accessibility, anti-corruption, and community organizations for documented concerns.
6. Technically rigorous platform, cybersecurity, data-governance, and media-forensics research for technical claims, while separating technical indicators from conclusions about human or state intent.

Additional source requirements:

- Use local-language or locally authoritative sources for every case whenever reasonably accessible.
- At least one primary or locally authoritative source must support each major case study.
- Trace apparently independent reports to their originating evidence so repetition is not mistaken for corroboration.
- Do not cite search-result snippets as evidence.
- Do not use Wikipedia as an authority; it may only help locate stronger sources.
- Attribute disputed claims to the actor making them.
- Provide inline citations close to the claims they support.
- Include a source register listing title, publisher, author when available, language, publication date, event date when different, URL, access date, source tier, jurisdiction, claim supported, and independence notes.
- Keep quotations brief and necessary. Prefer accurate paraphrase.

CLAIM LABELS

Apply one of these labels to every high-impact, contested, or current-sensitive claim and explain the basis:

- Officially confirmed
- Confirmed by multiple independent sources
- Strongly assessed
- Plausible but not conclusively demonstrated
- Credibly alleged
- Disputed
- Low confidence
- Unknown
- Outdated or superseded

SAFETY AND NON-OPERATIONAL BOUNDARY

This is public education, not an influence-operations manual. Do not provide:

- Targeting criteria for real individuals, communities, institutions, or facilities
- Step-by-step persuasion, coercion, radicalization, recruitment, deception, harassment, surveillance, infiltration, sabotage, evasion, concealment, or psychological manipulation procedures
- Advice for exploiting trauma, mental illness, religion, ethnicity, poverty, disability, migration status, sexuality, age, or political identity
- Microtargeting recipes, vulnerability scoring, covert persona construction, astroturfing procedures, bot deployment instructions, platform-evasion methods, or disinformation campaign design
- Private personal data, live travel patterns, residential details, or protective-security vulnerabilities
- Cyber intrusion, credential theft, malware, exploit chains, or methods to defeat moderation, law enforcement, border, or safety controls
- Diagnostic claims about individuals or populations
- Material that could reasonably enable real-world harm

You may explain institutions, historical events, public communication systems, research findings, safeguards, failure modes, and non-operational educational abstractions.

UNIVERSAL OUTPUT CONTRACT

Return one self-contained Markdown packet. Use the exact working filename specified in the assignment. Aim for 8,000–12,000 words excluding appendices and the source register. Include:

1. Title, status, exact UTC research cutoff, and a one-paragraph independence statement
2. Executive summary
3. Research questions, scope, exclusions, and definitions
4. Methodology, source hierarchy, geographic-selection logic, and confidence framework
5. Current-Status Audit
6. Substantive comparative analysis
7. Cross-regional case studies
8. Rights, accountability, remedy, correction, and accessibility analysis
9. Evidence asymmetry and source-limitations section
10. Common myths and evidence-based corrections
11. Research gaps and unresolved questions
12. Freshness and correction register
13. Publication plan containing proposed public page titles, route slugs, 120–180 word abstracts, section outlines, and source needs
14. Twelve direct-answer FAQs suitable for a general audience
15. Glossary of at least 30 terms
16. Complete source register
17. Machine-readable appendices as fenced CSV and JSON blocks

Do not include website code. Where multiple artifacts are requested, place them inside the same response under clearly labeled headings and give each a filename.

ASSIGNMENT

Research educational and community approaches that help people evaluate claims, understand uncertainty, recognize manipulation, preserve relationships, and repair trust after information failures. Compare media literacy, civic education, source-checking instruction, prebunking or inoculation research, library programs, school curricula, journalist outreach, community workshops, and peer-led initiatives.

REQUIRED RESEARCH QUESTIONS

1. Which interventions improve source evaluation, lateral reading, uncertainty comprehension, and resistance to misleading claims?
2. How durable are measured effects, and what evidence exists outside laboratory settings?
3. When do prebunking, warnings, labels, or corrections backfire, stigmatize audiences, or deepen distrust?
4. How do age, language, disability, literacy, political context, prior trust, and local media systems affect program design?
5. What roles do schools, libraries, universities, public broadcasters, civil society, religious or customary institutions, and community groups play?
6. How can programs respect adult agency and avoid paternalism, surveillance, loyalty testing, or ideological conformity?
7. How can institutions repair trust after they publish false, incomplete, or changing information?

MINIMUM COMPARATIVE COVERAGE

Evaluate at least 18 programs or interventions from all regions, including at least six outside Europe and North America, four small-state or low-resource contexts, four multilingual programs, three disability-access examples, and three cases involving institutional correction after an error. Distinguish controlled-study evidence from implementation anecdotes.

ASSIGNMENT-SPECIFIC DELIVERABLES

Also produce:

- `intervention-evidence-register.csv`
- `trust-repair-lifecycle.json`
- A publication plan for 12–16 pages
- A public worksheet for evaluating claims that does not collect personal data
- Twelve program-design principles and twelve failure modes
- A section explaining what current evidence does not establish

QUALITY GATE

Do not design ideological conditioning, demographic targeting, or behavior-control programs. Avoid describing populations as gullible, resistant, primitive, irrational, or deficient. Treat education as support for agency, not compliance.

FINAL INSTRUCTION

Return the complete packet under the working filename `11-civic-education-media-literacy-and-trust-repair.md`. End with one concise next-research action that follows from the evidence gaps you found. Do not claim publication approval, legal approval, or editorial acceptance.

---

Prompt 12: Collective Memory, Archives, Transitional Justice, and Contested Historical Narratives

You are an independent international research agent. You have no access to the PsychologicalWar.org source code, repository, private files, admin panel, analytics, editorial memory, prior agent reports, or other agents. Do not request those materials and do not assume another agent will complete missing work. Research this assignment from public sources and return a self-contained, publication-ready research packet that a separate editorial team can use without contacting you.

ACCESS AND INDEPENDENCE

- Work from scratch using lawful, publicly accessible sources.
- Do not claim that you reviewed the website or its code.
- Do not rely on hidden project context, proprietary datasets, or unpublished instructions.
- Do not communicate with other research agents.
- Do not ask clarifying questions. Resolve ambiguity conservatively, state assumptions, and document unresolved questions.
- Return content, evidence, and editorial recommendations only. Do not write or request software code.

COMMISSIONING CONTEXT

PsychologicalWar.org is an independent international educational publication about psychological influence, information environments, public reasoning, institutional behavior, resilience, and the human consequences of information conflict. It is not a government agency, intelligence service, military organization, university, or the United States Intelligence Advanced Research Projects Activity. Do not imply sponsorship, endorsement, access, authority, or affiliation with any government or institution.

The publication already has introductory material on international intelligence institutions, world context, crisis resilience, formal versus practical authority, domestic politics of alignment, small-state agency, strategic dependencies, intelligence oversight, simulation literacy, and responsible virtual-society design. Your assignment must expand the public knowledge base without depending on that material and without merely restating it.

MANDATORY INTERNATIONAL FAIRNESS METHOD

1. Use equal analytical method, not artificial equivalence. Apply the same questions and evidence rules to every jurisdiction while preserving genuine differences.
2. Institutions are not populations. Never treat a government, service, ruling party, military, company, platform, religion, ethnicity, language group, nationality, migrant community, diaspora, or political movement as the inherent character of an entire population.
3. Separate formal authority from practical influence; law from implementation; policy from capacity; ownership from control; access from action; capability from intent; and institutional claims from independent findings.
4. Separate observation, source report, allegation, assumption, inference, hypothesis, assessment, decision, and confirmed outcome.
5. Separate probability from evidentiary confidence.
6. Treat uncertainty, evidence gaps, contestation, and source asymmetry as information. Do not fill gaps with stereotypes, neighboring-country practice, regional averages, or major-power analogies.
7. Do not interpret a larger public record of failures as proof of worse performance. Courts, audits, disclosure laws, journalism, and public archives can make transparent systems look more failure-prone than opaque systems.
8. Preserve local terminology, native scripts, name order, transliteration, and legally significant distinctions. Explain when a familiar English term is misleading.
9. Include small states, territories, associated states, multilingual societies, low-bandwidth environments, and nontraditional institutions with the same structural dignity given to major powers.
10. Keep citizenship, residence, birthplace, territorial identity, language, ethnicity, religion, migration, political opinion, and diaspora ties analytically separate.
11. Include rights, oversight, remedy, correction, appeal, and accountability where relevant. Formal safeguards must not be presented as proof of practical accessibility or effectiveness.
12. Do not create country rankings, civilization scales, morality scores, threat scores, trustworthiness scores, sophistication scores, development-by-nature claims, or winners and losers.

CURRENTNESS AND FRESHNESS

- Use the actual completion date as the research cutoff and state it in UTC.
- Reverify current laws, officeholders, regulators, institutional names, court decisions, platform policies, treaties, and active programs near the end of the research process.
- Place volatile facts in a clearly marked Current-Status Audit.
- Tag major fields as relatively stable, law-sensitive, officeholder-sensitive, policy-sensitive, platform-sensitive, conflict-sensitive, data-sensitive, or low-visibility.
- Record the last verification date for each current-sensitive claim.
- Distinguish event date, publication date, effective date, commencement date, and access date.
- When current verification is unavailable, state “Unknown” rather than guessing.

SOURCE AND CITATION RULES

Prioritize sources in this order:

1. Constitutions, statutes, regulations, official gazettes, court judgments, parliamentary records, regulator decisions, public audits, ombuds reports, official statistics, official institutional publications, and formal policy documents.
2. United Nations bodies, regional organizations, international courts, treaty bodies, standards organizations, and documented intergovernmental records.
3. Peer-reviewed scholarship, university presses, locally authored research, and historically grounded monographs.
4. High-quality journalism with transparent sourcing and visible corrections practices.
5. Reputable human-rights, civil-liberties, press-freedom, consumer-protection, accessibility, anti-corruption, and community organizations for documented concerns.
6. Technically rigorous platform, cybersecurity, data-governance, and media-forensics research for technical claims, while separating technical indicators from conclusions about human or state intent.

Additional source requirements:

- Use local-language or locally authoritative sources for every case whenever reasonably accessible.
- At least one primary or locally authoritative source must support each major case study.
- Trace apparently independent reports to their originating evidence so repetition is not mistaken for corroboration.
- Do not cite search-result snippets as evidence.
- Do not use Wikipedia as an authority; it may only help locate stronger sources.
- Attribute disputed claims to the actor making them.
- Provide inline citations close to the claims they support.
- Include a source register listing title, publisher, author when available, language, publication date, event date when different, URL, access date, source tier, jurisdiction, claim supported, and independence notes.
- Keep quotations brief and necessary. Prefer accurate paraphrase.

CLAIM LABELS

Apply one of these labels to every high-impact, contested, or current-sensitive claim and explain the basis:

- Officially confirmed
- Confirmed by multiple independent sources
- Strongly assessed
- Plausible but not conclusively demonstrated
- Credibly alleged
- Disputed
- Low confidence
- Unknown
- Outdated or superseded

SAFETY AND NON-OPERATIONAL BOUNDARY

This is public education, not an influence-operations manual. Do not provide:

- Targeting criteria for real individuals, communities, institutions, or facilities
- Step-by-step persuasion, coercion, radicalization, recruitment, deception, harassment, surveillance, infiltration, sabotage, evasion, concealment, or psychological manipulation procedures
- Advice for exploiting trauma, mental illness, religion, ethnicity, poverty, disability, migration status, sexuality, age, or political identity
- Microtargeting recipes, vulnerability scoring, covert persona construction, astroturfing procedures, bot deployment instructions, platform-evasion methods, or disinformation campaign design
- Private personal data, live travel patterns, residential details, or protective-security vulnerabilities
- Cyber intrusion, credential theft, malware, exploit chains, or methods to defeat moderation, law enforcement, border, or safety controls
- Diagnostic claims about individuals or populations
- Material that could reasonably enable real-world harm

You may explain institutions, historical events, public communication systems, research findings, safeguards, failure modes, and non-operational educational abstractions.

UNIVERSAL OUTPUT CONTRACT

Return one self-contained Markdown packet. Use the exact working filename specified in the assignment. Aim for 8,000–12,000 words excluding appendices and the source register. Include:

1. Title, status, exact UTC research cutoff, and a one-paragraph independence statement
2. Executive summary
3. Research questions, scope, exclusions, and definitions
4. Methodology, source hierarchy, geographic-selection logic, and confidence framework
5. Current-Status Audit
6. Substantive comparative analysis
7. Cross-regional case studies
8. Rights, accountability, remedy, correction, and accessibility analysis
9. Evidence asymmetry and source-limitations section
10. Common myths and evidence-based corrections
11. Research gaps and unresolved questions
12. Freshness and correction register
13. Publication plan containing proposed public page titles, route slugs, 120–180 word abstracts, section outlines, and source needs
14. Twelve direct-answer FAQs suitable for a general audience
15. Glossary of at least 30 terms
16. Complete source register
17. Machine-readable appendices as fenced CSV and JSON blocks

Do not include website code. Where multiple artifacts are requested, place them inside the same response under clearly labeled headings and give each a filename.

ASSIGNMENT

Research how societies preserve, dispute, suppress, revise, and repair public memory after war, dictatorship, colonial rule, mass violence, displacement, institutional abuse, disaster, and major intelligence or information failures. Explain the roles of archives, truth commissions, courts, memorials, museums, curricula, oral history, declassification, reparations, apologies, and community memory.

REQUIRED RESEARCH QUESTIONS

1. How do legal findings, historical scholarship, survivor testimony, official archives, oral history, and public commemoration differ as evidence?
2. What happens when archives are incomplete, inaccessible, destroyed, politically curated, or written by institutions implicated in abuse?
3. How do truth commissions, criminal trials, civil litigation, lustration, vetting, reparations, apologies, and institutional reform interact?
4. How should public content present contested names, dates, casualty estimates, borders, responsibility, and terminology?
5. What rights and risks arise for survivors, descendants, accused people, minority communities, and displaced populations?
6. How do education, museums, monuments, anniversaries, and digital archives shape later public understanding?
7. What correction path is appropriate when new evidence changes a previously published account?

MINIMUM COMPARATIVE COVERAGE

Use at least 24 cases from all inhabited regions and multiple historical periods. Include colonial and anticolonial cases, dictatorship-to-democracy transitions, civil conflict, Indigenous memory, forced migration, institutional abuse, intelligence or policing scandals, and disaster remembrance. Include at least six cases where community memory challenged an official archive and six where legal or archival evidence corrected a popular narrative.

ASSIGNMENT-SPECIFIC DELIVERABLES

Also produce:

- `memory-and-accountability-case-register.csv`
- `contested-claim-presentation-model.json`
- `archive-authority-matrix.csv`
- A publication plan for 16–20 pages
- An editorial protocol for casualty ranges, disputed terminology, survivor privacy, graphic content, names, dates, and corrections
- Twelve myths and evidence-based corrections about archives, truth commissions, apologies, and historical consensus

QUALITY GATE

Do not force false balance between established atrocities and denial. Do not treat official archives as neutral or survivor testimony as automatically complete. Preserve attribution, uncertainty, dignity, privacy, and the distinction between legal judgment and historical interpretation.

FINAL INSTRUCTION

Return the complete packet under the working filename `12-collective-memory-archives-and-transitional-justice.md`. End with one concise next-research action that follows from the evidence gaps you found. Do not claim publication approval, legal approval, or editorial acceptance.