context-engineering

Pass

Audited by Gen Agent Trust Hub on May 9, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns detected. The skill provides clear guidance on managing Claude Code context limits using official mechanisms.
  • [COMMAND_EXECUTION]: The instructions suggest using standard Claude Code commands like /compact and /tokens, which are intended platform features.
  • [PERSISTENCE]: The documentation mentions setting environment variables (e.g., CLAUDE_AUTOCOMPACT_PCT_OVERRIDE) in shell profiles (~/.bashrc, ~/.zshrc). This is a standard and legitimate configuration practice for environment-based settings and does not constitute a malicious persistence mechanism.
  • [DATA_EXFILTRATION]: The skill recommends using local files within the project's .claude/ directory to store session state and analysis results. No network transmission to external or untrusted domains was found.
Audit Metadata
Risk Level
SAFE
Analyzed
May 9, 2026, 08:40 PM
Security Audit — agent-trust-hub — context-engineering