skill-cleaner

Pass

Audited by Gen Agent Trust Hub on Jul 15, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill requires running a provided TypeScript script (skill-cleaner.ts) using the Node.js runtime to perform the audit.
  • [DATA_EXFILTRATION]: The auditing script reads user configuration files, model caches, and session logs (e.g., ~/.pi/sessions, ~/.codex/history.jsonl) to identify usage patterns. This behavior exposes user history and system configuration to the agent's context as part of the audit report.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from third-party SKILL.md files and session logs, creating a potential surface for indirect injection if malicious content within those files is included in the summary report.
  • Ingestion points: Scanning of third-party directories (git/, npm/) and session log files.
  • Boundary markers: The output is structured as a Markdown report; no explicit 'ignore instructions' markers are present for the aggregated content.
  • Capability inventory: The script performs recursive directory scanning and local file reading.
  • Sanitization: The script includes a sanitizeSingleLine function to flatten descriptions and remove newlines, reducing the likelihood of control flow interruption.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 15, 2026, 05:09 PM
Security Audit — agent-trust-hub — skill-cleaner