Back to skills
SkillHub ClubShip Full StackFull Stack

evidence-auditor

Audit the evidence supporting each claim and write gaps/concerns into `output/MISSING_EVIDENCE.md`. **Trigger**: evidence audit, missing evidence, unsupported claims, 审稿证据审计, 证据缺口. **Use when**: peer review 流程中,需要逐条检查 claim 的证据链、缺 baseline、评测薄弱点。 **Skip if**: 缺少 claims 输入(例如还没有 `output/CLAIMS.md`)。 **Network**: none. **Guardrail**: 只写“缺口/风险/下一步验证”,不要替作者补写论述或引入新主张。

Packaged view

This page reorganizes the original catalog entry around fit, installability, and workflow context first. The original raw source lives below.

Stars
334
Hot score
99
Updated
March 20, 2026
Overall rating
C4.3
Composite score
4.3
Best-practice grade
A88.4

Install command

npx @skill-hub/cli install willoscar-research-units-pipeline-skills-evidence-auditor

Repository

WILLOSCAR/research-units-pipeline-skills

Skill path: .codex/skills/evidence-auditor

Audit the evidence supporting each claim and write gaps/concerns into `output/MISSING_EVIDENCE.md`. **Trigger**: evidence audit, missing evidence, unsupported claims, 审稿证据审计, 证据缺口. **Use when**: peer review 流程中,需要逐条检查 claim 的证据链、缺 baseline、评测薄弱点。 **Skip if**: 缺少 claims 输入(例如还没有 `output/CLAIMS.md`)。 **Network**: none. **Guardrail**: 只写“缺口/风险/下一步验证”,不要替作者补写论述或引入新主张。

Open repository

Best for

Primary workflow: Ship Full Stack.

Technical facets: Full Stack.

Target audience: everyone.

License: Unknown.

Original source

Catalog source: SkillHub Club.

Repository owner: WILLOSCAR.

This is still a mirrored public skill entry. Review the repository before installing into production workflows.

What it helps with

  • Install evidence-auditor into Claude Code, Codex CLI, Gemini CLI, or OpenCode workflows
  • Review https://github.com/WILLOSCAR/research-units-pipeline-skills before adding evidence-auditor to shared team environments
  • Use evidence-auditor for development workflows

Works across

Claude CodeCodex CLIGemini CLIOpenCode

Favorites: 0.

Sub-skills: 0.

Aggregator: No.

Original source / Raw SKILL.md

---
name: evidence-auditor
description: |
  Audit the evidence supporting each claim and write gaps/concerns into `output/MISSING_EVIDENCE.md`.
  **Trigger**: evidence audit, missing evidence, unsupported claims, 审稿证据审计, 证据缺口.
  **Use when**: peer review 流程中,需要逐条检查 claim 的证据链、缺 baseline、评测薄弱点。
  **Skip if**: 缺少 claims 输入(例如还没有 `output/CLAIMS.md`)。
  **Network**: none.
  **Guardrail**: 只写“缺口/风险/下一步验证”,不要替作者补写论述或引入新主张。
---

# Evidence Auditor (peer review)

Goal: for each claim, either (a) point to the supporting evidence in the manuscript, or (b) write a concrete gap with an actionable fix.

## Inputs

- `output/CLAIMS.md`

## Outputs

- `output/MISSING_EVIDENCE.md`

## Output format (recommended)

For each claim:
- `Claim`: copy the claim text
- `Evidence present`: what the paper provides (experiments/theory/citations)
- `Gap / concern`: what is missing or weak
- `Minimal fix`: the smallest additional evidence that would address the gap
- `Severity`: `major` | `minor` (optional)

## Workflow

1. Iterate claims in `output/CLAIMS.md`.
2. For empirical claims, check:
   - dataset/task definition is clear
   - baselines are appropriate
   - evaluation protocol is valid
   - ablations/sensitivity analyses exist where needed
3. For conceptual claims, check:
   - definitions are unambiguous
   - assumptions are stated
   - claims do not exceed what is argued
4. Write `output/MISSING_EVIDENCE.md` as a list of claim-by-claim entries.

## Definition of Done

- [ ] Every claim from `output/CLAIMS.md` has an evidence note or a gap item.
- [ ] “Fix” items are actionable (what to add, not “more experiments”).

## Troubleshooting

### Issue: you cannot locate the evidence in the paper

**Fix**:
- Mark the claim as “evidence not locatable” and ask for a clearer source pointer (or re-extract claims with better pointers).

### Issue: the audit starts proposing new claims

**Fix**:
- Stop; only critique what exists in `output/CLAIMS.md` and the manuscript.