Skip to content
Get packs

Integrity Forensics Audit

Runs a pinned integrity audit on a paper and turns the findings into a block/warn gate

Source Download skill
Repository
wanshuiyin/Auto-claude-code-research-in-sleep
Stars
18,482
Licence
MIT
Files
1 file

What it does

When you ask the agent to forensically check a paper for integrity problems, this skill makes it run a fixed, version-pinned external audit tool (Anti-Autoresearch) against the paper. The audit uses automated statistical checks plus GPT-based reviewer agents to list possible issues, then converts the result into a simple verdict (BLOCK, WARN, or NO_NEW_BLOCKER) and an append-only record of follow-up obligations. The skill itself is a launcher that clones and runs that upstream tool rather than doing the analysis on its own.

Great for

  • Authors doing a self-check of a manuscript before submission
  • Reviewers who want an integrity appendix for a review
  • Researchers who already use the ARIS pipeline and want a standard gate

Skip it if

  • Final judgements of misconduct: the tool summarizes evidence and does not adjudicate, so a flag is a prompt for human review, not proof
  • Users without git, Python 3 and access to the Codex/GPT tool, since the audit cannot run without them
  • Iterating edits until the tool stops flagging, which the skill itself forbids because it would game the verdict

Try asking

  • “Run integrity forensics on ./my-paper before I submit it”
  • “Forensic audit this paper: arxiv 2401.00001”
  • “投稿前自查诚信 for my manuscript PDF”
  • “Check this paper's statistics with anti-autoresearch and give me the BLOCK/WARN verdict”
Step-by-step guide: how to use Integrity Forensics AuditInstall, first run, examples, tips and FAQs

Install in 3 steps

  1. 1

    Download the skill

    The ZIP contains the 1 original files, the MIT licence and an attribution note.

    integrity-forensics.zip
  2. 2

    Add it to your agent

    Claude Code (all projects)

    mkdir -p ~/.claude/skills && unzip integrity-forensics.zip -d ~/.claude/skills/

    Claude Code / Agent SDK (this project only)

    mkdir -p .claude/skills && unzip integrity-forensics.zip -d .claude/skills/

    Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.

  3. 3

    Ask for the task

    No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.

Get better results

  • Let the pinned upstream eval gate pass once; the skill refuses to produce verdicts if it fails
  • Treat flags as items to check by hand, and read the auditor's stated reasoning in the report
  • Provide the paper by absolute path or a clear directory so the audit does not resolve the wrong repository

What's inside

  • SKILL.md

Redistributed unmodified from wanshuiyin/Auto-claude-code-research-in-sleep under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.