Skip to content

chore: add nucleus security scan for Claude settings#1578

Closed
brandon-coproduct wants to merge 6 commits intolark-parser:masterfrom
coproduct-opensource:add-nucleus-scan
Closed

chore: add nucleus security scan for Claude settings#1578
brandon-coproduct wants to merge 6 commits intolark-parser:masterfrom
coproduct-opensource:add-nucleus-scan

Conversation

@brandon-coproduct
Copy link

Summary

  • Adds .claude/settings.json with restrictive agent permissions
  • Adds nucleus-scan.yml workflow to scan Claude settings on PRs

The nucleus scan action performs deterministic security analysis of AI agent configurations — no LLM required. It checks for the lethal trifecta (private data + untrusted content + exfiltration) and other security issues.

Test plan

  • Nucleus scan workflow passes on this PR
  • Verified locally: nucleus-audit scan --claude-settings .claude/settings.json → PASS

🤖 Generated with Claude Code

brandon-coproduct and others added 6 commits March 7, 2026 03:08
fix: pin nucleus v1.0.2 + github-token for PR creation
Adds `.claude/settings.json` with restrictive permissions:
- Read-only access + python/pip/pytest/git commands allowed
- Exfiltration commands denied (curl, wget, nc, ssh, scp)
- Sensitive paths denied (.env, credentials, secrets)
- Write/edit/commit require approval

Adds `nucleus-scan.yml` workflow that scans the Claude settings
on every PR that modifies agent config files.

Generated with [Claude Code](https://claude.ai/code)
via [Happy](https://happy.engineering)

Co-Authored-By: Claude <noreply@anthropic.com>
Co-Authored-By: Happy <yesreply@happy.engineering>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants