docs-seeker

Pass

Audited by Gen Agent Trust Hub on Jun 26, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill performs downloads from well-known sources including documentation websites, GitHub repositories, and the npm registry (for the Repomix tool). These downloads are essential for the primary purpose of technical documentation analysis.
  • [COMMAND_EXECUTION]: Uses shell commands to clone repositories and execute the Repomix analysis tool. The skill safely uses the /tmp/ directory for these operations.
  • [PROMPT_INJECTION]: The skill includes a surface for indirect prompt injection as it ingests and processes untrusted documentation and repository data.
  • Ingestion points: External documentation URLs and GitHub repository files.
  • Boundary markers: No specific delimiters are used to wrap external content, although the skill emphasizes source verification.
  • Capability inventory: Shell execution (git, repomix) and sub-agent coordination via the Task tool.
  • Sanitization: The skill focuses on data aggregation and does not implement specific sanitization logic for fetched content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 26, 2026, 06:52 PM
Security Audit — agent-trust-hub — docs-seeker