claim-repro-score
One repeatable RSI pass over claim reproducibility; run the claim-repro scorecard, retire unfalsifiable witness debt worst-first by adding the real artifact/test/path, re-measure, and commit only the claim-repro lane. Use when this named workflow matches the task.
- License 许可
- Apache-2.0
- 最近更新
- 2026-11-15
- Stars
- ★ 48
- Forks
- 16
01 前言 Front Matter
- name
- claim-repro-score
- metadata
- { "canonical": "../../../.claude/skills/claim-repro-score/SKILL.md", "generated-by": "fak project-assets sync" }
- description
- One repeatable RSI pass over claim reproducibility; run the claim-repro scorecard, retire unfalsifiable witness debt worst-first by adding the real artifact/test/path, re-measure, and commit only the claim-repro lane. Use when this named workflow matches the task.
03 SKILL.md 正文
One repeatable RSI pass over claim reproducibility — the truth-maintenance sister to conflation-score (who owns this number?) and appeal-score (prose voice). Runs the claim-repro scorecard (tools/claim_repro_scorecard.py) over CLAIMS.md and BENCHMARK-AUTHORITY.md, reads the claim-repro-debt work-list (un-falsifiable witnesses: missing artifacts, non-existent test patterns, deleted cmd dirs), retires debt worst-first by ADDING the real witness (write the missing artifact, add the missing test, correct the package path), re-measures to PROVE the debt dropped, and commits only the claim-repro lane by explicit path. The claim-falsifiability sister of refresh-readme (freshness) and slop-score (code waste). Use after adding a claim or benchmark, when a witness is deleted or renamed, or on a /loop cadence to keep every claim genuinely falsifiable from a clean clone.
04 同仓库更多 Skills
serve-receipt
On-demand research protocol + deterministic receipt for serving a Hugging Face model on a concrete GPU target — fan out research lanes (model facts/files, engine support matrix, serving recipes, KV/state memory math), compute the tiered feasibility verdict with `fak serve-receipt` (never narrated math), and emit a witnessed launch receipt. Use when asked for an "initial config", "serving receipt", or "how do I serve <model> on <GPUs>".
caveman
Ultra-compressed communication mode. Cuts output tokens 65% (measured) by speaking like caveman while keeping full technical accuracy. Supports intensity levels: lite, full (default), ultra, wenyan-lite, wenyan-full, wenyan-ultra. Use when user says "caveman mode", "talk like caveman", "use caveman", "less tokens", "be brief", or invokes /caveman. Also auto-triggers when token efficiency is requested.
quality-report
Score the repo quality and emit a report card.
ctx-overlay
Compact the context window and prune the memory store when a session grows too long.
quality-report
Score the repo quality and emit a report card.
ctx-overlay
Compact the context window and prune the memory store when a session grows too long.