~ / rigs / wanshuiyin / Anti-Autoresearch

wanshuiyin/Anti-Autoresearch

Don't trust an autoresearch paper at face value. Reviewer-side integrity forensics (self-consistency + fabrication), deterministic verdict. 61 signals: 46 integrity hack-patterns (families A–H, verdict-bearing) + 13 zero-weight AI writing-s

↗ GitHub ★ 159 mit updated 11d ago project Claude Code
share on X
ARCHETYPE
Skill Collector
Ten-plus skills loaded on demand. A procedural-knowledge library.
CONTEXT TAX · EVERY TURN
~969 tokens
Featherweight · median rig: 2.3k · breakdown
GUARDRAILS
0/5
No committed guardrails · details

Copy this rig

# review before running: this installs third-party code
$ npx degit wanshuiyin/Anti-Autoresearch/skills ./rig-anti-autoresearch/skills

MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.

$ npx degit wanshuiyin/Anti-Autoresearch/skills/adversarial-case-builder .claude/skills/adversarial-case-builder
$ npx degit wanshuiyin/Anti-Autoresearch/skills/ai-style-impressions .claude/skills/ai-style-impressions
$ npx degit wanshuiyin/Anti-Autoresearch/skills/baseline-comparison-audit .claude/skills/baseline-comparison-audit
$ npx degit wanshuiyin/Anti-Autoresearch/skills/citation-forensics .claude/skills/citation-forensics
$ npx degit wanshuiyin/Anti-Autoresearch/skills/consistency-audit .claude/skills/consistency-audit
$ npx degit wanshuiyin/Anti-Autoresearch/skills/eval-design-forensics .claude/skills/eval-design-forensics
$ npx degit wanshuiyin/Anti-Autoresearch/skills/evidence-ledger .claude/skills/evidence-ledger
$ npx degit wanshuiyin/Anti-Autoresearch/skills/experiment-forensics .claude/skills/experiment-forensics
$ npx degit wanshuiyin/Anti-Autoresearch/skills/novelty-duplication-advisory .claude/skills/novelty-duplication-advisory
$ npx degit wanshuiyin/Anti-Autoresearch/skills/presentation-signals .claude/skills/presentation-signals
$ npx degit wanshuiyin/Anti-Autoresearch/skills/proof-derivation-forensics .claude/skills/proof-derivation-forensics

This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 7,204 rigs:

{
  "permissions": {
    "deny": [
      "Read(./.env)",
      "Read(**/.env)",
      "Read(~/.ssh/**)",
      "Bash(rm -rf *)",
      "Read(**/*.pem)",
      "Bash(rm -rf /)",
      "Bash(git push --force:*)",
      "Bash(sudo *)",
      "Read(.env)",
      "Bash(rm -rf /*)",
      "Read(./.env.*)",
      "Read(~/.aws/**)",
      "Bash(git push --force*)",
      "Bash(rm -rf:*)",
      "Read(**/*.key)",
      "Read(**/.env.*)",
      "Bash(sudo:*)",
      "Bash(git reset --hard*)",
      "Bash(git reset --hard:*)",
      "Read(.env.*)"
    ],
    "ask": [
      "Bash(git push:*)",
      "Bash(git push *)",
      "Bash(git commit:*)",
      "Bash(rm *)",
      "Bash(rm:*)",
      "Bash(git rebase *)",
      "Bash(wget *)",
      "Bash(npm publish:*)",
      "Bash(git commit *)",
      "Bash(gh pr merge *)"
    ]
  }
}

Skills (11)

Similar rigs

copied ✓