← ClaudeAtlas

mizanlisted

Evidence-tiered claim auditing and preregistered hypothesis registries for documents, codebases, bugs, and features. Use whenever the user asks to evaluate, audit, review, or fact-check any claim set (AI-generated summaries, project reports, self-assessments, research writeups, year-in-reviews), wants honest rigor instead of praise, or mentions evidence tiers, preregistration, refutation conditions, HARKing, confounds, or hit rates. Also use to start or maintain hypothesis registries for experiments, decisions, or predictions. Also use for software work — auditing an existing codebase or repo (what comments, names, docs, and tests claim vs. what code does; generating evidence-tiered docs from undocumented code), tracking bug hypotheses while debugging, and gating new features or PRDs (tiering PRD claims, preregistering success metrics and kill conditions, generating alternatives from gap maps). Triggers include "değerlendir", "denetle", "önkayıt", "audit this repo", "bug hipotezi", "PRD'yi süz".
XINMurat/Mizan · ★ 1 · Data & Documents · score 65
Install: claude install-skill XINMurat/Mizan
# Mizan — Evidence-Tiered Auditing & Preregistration Registry Mizan (Turkish/Arabic: "the scale") turns a rigorous experimental-science discipline into a portable tool for evaluating *any* claim set and for maintaining living hypothesis registries. Its core commitments: 1. **Every claim gets an evidence tier.** No untagged assertions. 2. **Thresholds are locked before results.** If that's impossible (retrospective analysis), the HARKing risk is stated explicitly — never silently absorbed. 3. **Every hypothesis carries a refutation condition.** A claim that cannot fail is not audited, it is decorated. 4. **Refuted entries are never deleted.** They are marked `[R]` and archived in place. Negative results are first-class results. 5. **Surprising positives get a symmetric control before a headline.** A result that flatters the hypothesis is the one most in need of a confound check. 6. **Hit rates over curated examples.** Three confirming anecdotes are selection bias; a scored prediction record is evidence. ## Evidence tiers (use these exact labels, bilingual) | Tag | TR | EN | Meaning | |---|---|---|---| | `[K]` | Kanıtlanmış | Proven | Direct evidence supports it; source cited; threshold met | | `[H]` | Makul Hipotez | Plausible hypothesis | Theoretical grounding exists; empirical support missing or below threshold | | `[S]` | Spekülatif | Speculative | Interesting; not currently testable or no test designed | | `[R]` | Reddedildi | Refuted | Tested and