auto-claude

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill documentation follows security best practices for credential management by instructing users to store sensitive tokens (like CLAUDE_CODE_OAUTH_TOKEN and GOOGLE_API_KEY) in local environment files (.env) rather than hardcoding them within the skill instructions or scripts.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and act upon user-provided feature descriptions and existing codebase contents. While this represents a standard attack surface for autonomous coding tools, the risk is inherent to the primary purpose of the skill and no specific vulnerabilities were identified.
  • Ingestion points: The skill processes user input via the /auto-claude command and performs semantic analysis on local codebase files.
  • Boundary markers: Not specified in the provided documentation.
  • Capability inventory: The skill manages git worktrees and performs autonomous file system writes/modifications via multi-agent coordination.
  • Sanitization: Not specified in the provided documentation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 08:36 PM
Security Audit — agent-trust-hub — auto-claude