strategic-compact

Pass

Audited by Gen Agent Trust Hub on Aug 15, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses a shell script (suggest-compact.sh) intended to run as a hook during tool use (e.g., Edit/Write operations). This is a documented platform feature for augmenting agent behavior.
  • [DATA_EXPOSURE]: The script creates a temporary file in /tmp/ to track session tool counts. It uses CLAUDE_SESSION_ID to ensure session isolation, preventing cross-session data leakage.
  • [REMOTE_CODE_EXECUTION]: The skill references an external URL for a 'Longform Guide' on X (Twitter). This is a standard informational link and does not involve automated downloads or execution.
  • [SAFE_PRACTICE]: The instructions emphasize user-initiated action (/compact) rather than automated state changes, ensuring the user remains in control of the agent's context.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 15, 2026, 03:23 AM
Security Audit — agent-trust-hub — strategic-compact