Project Explorer

Files

Story files

Browse the notes, research, decisions, and workshops used to develop the series.

Search 577 documents about the world, characters, plot, research, decisions, workshops, and earlier ideas.

Current document skills/research-story-material/SKILL.md

Document metadata
name: research-story-material
description: Research real-world knowledge that can inform Seeds of the Throne without inventing story facts or converting evidence into canon. Use for scientific, historical, psychological, technological, political, legal, medical, cultural, or institutional plausibility questions, research requests, source evaluation, reports, and reusable findings.

Research Story Material

Define the decision

Read 03 Context/RULES.md, 04 Research/README.md, and the minimum relevant story context. State the story or plausibility question the research should inform. Define exclusions before searching.

If another agent is being assigned research, create a request from 04 Research/Requests/REQUEST TEMPLATE.md and prohibit invention of canon, characters, factions, names, scenes, or plot.

Investigate

  1. Search current primary and authoritative sources first.
  2. Use secondary sources for synthesis or when primary evidence is unavailable.
  3. Separate established evidence, contested interpretation, plausible extrapolation, and speculation.
  4. Record constraints, failure modes, ethical limits, and useful terminology, not only supporting examples.
  5. Capture publication date, direct source link, and enough context to audit each important claim.
  6. Distinguish source claims from your own inference.

Use references/evidence-standard.md for source and claim handling.

Translate for story use

Extract mechanisms rather than copying real people, nations, firms, movements, or events. Offer implications as options the author may adopt, modify, or reject. State clearly that research does not establish canon.

Save results

  • Put the complete sourced report in 04 Research/Full Reports/.
  • Put each durable, concise takeaway in 04 Research/Findings/ and link its report.
  • Record whether an implication is adopted, rejected, or open only after an author decision.
  • Do not update 02 Story/ merely because research supports an idea.

Guardrails

  • Do not fabricate citations or use unverifiable quotations.
  • Do not disguise uncertainty with confident language.
  • Do not diagnose characters or ground villainy in identity or mental illness.
  • Do not introduce real-person analogues into compiled story data.
  • Browse when facts may have changed, accuracy is consequential, or precise attribution is needed.