canon-rules
Skill2 starsChanged 7 months agoMITView on GitHub
Reference catalog of canon enforcement principles, epistemological operating rules, metascience diagnostic filters, adversarial validation protocols, and anti-patterns to avoid. Use as a reference when auditing fails, when a new idea conflicts with existing canon, when diagnosing inconsistency, or when user says "anti-patterns", "epistemology", "canon rules", "reasoning principles", "how do I check this", "what am I missing", "why is this inconsistent", "grill me harder", "adversarial review", "red team this", or "Russell Conjugation". Does NOT execute audits — use /auditing-physics or /auditing-canon for that.
In this repository19 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- auditing-canonSkill
- auditing-human-assumptionsSkill
- auditing-physicsSkill
- compiling-entitiesSkill
- cross-checkingSkill
- deep-readingSkill
- deriving-social-systemsSkill
- designing-loreSkill
- designing-worldsSkill
- extracting-entitiesSkill
- integrating-researchSkill
- interrogating-designSkill
- planning-worldbuildingSkill
- researching-papersSkill
- verifying-completionSkill
- writing-scienceSkill
- writing-styleSkill
- writing-worldbuildingSkill
The file
Files
Canon Rules — Epistemology, Adversarial Protocols & Anti-Patterns
Quick Reference — The Five Operating Principles
- Mechanism over association — state the causal chain or it's not proven
- First-principles over tropes — derive from physics/biology, not genre convention
- Evidence hierarchy — canon source > derived lore > cross-ref inference > "feels right"
- Anti-reductionism — trace implications through every connected system
- Adversarial self-validation — "what existing canon contradicts this?"
Detailed References (load on demand)
references/epistemological-framework.md— Full framework with named sources: Ioannidis corollaries, Garden of Forking Paths (Gelman), Aesthetic Fallacy (Hossenfelder), Humpty Dumpty Problem (Taleb), predictive power as ultimate metric (Broken Science Initiative), OODA Loop (Boyd)references/adversarial-protocols.md— Adversarial Red Teaming, Adversarial Collaboration (Kahneman protocol), Russell Conjugation stripping, Self-Correction Loop, Ensemble Thinking (Diversity Prediction Theorem)references/metascience-canon-filters.md— 6 diagnostic filters adapted from Ioannidis for worldbuilding: sample size, effect size, confirmation bias, flexible analysis, replication, survivorship bias
Anti-Patterns (Quick List)
- Vague Lore — generic without checkable detail
- Missing Locked Canon — settled decisions not recorded
- No Verification Step — declaring clear without full-repo grep
- Too Much in One Pass — mixing physics + prose audits
- Patching Forward — 3+ contradictory fixes → revert, rethink
- Flexible Analysis — reinterpreting a rule to accommodate new lore
- Trusting Without Location — no file + line = reject the finding
- The New Orthodoxy Trap — the system itself becoming rigid dogma (see Weinstein critique)
Out of Scope
This skill is a reference catalog. It does NOT execute audits (use /auditing-physics or /auditing-canon), stress-test decisions (use /interrogating-design), or check wiki structure (use /lint-wiki).
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.