parallel-agents
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- SAFE (SAFE): The skill consists entirely of instructional markdown without executable scripts, external dependencies, or binary files.
- DATA_EXPOSURE (SAFE): The 'allowed-tools' (Read, Glob, Grep) are limited to filesystem discovery and reading, which is consistent with the stated purpose of codebase analysis. No network tools are requested.
- INDIRECT_PROMPT_INJECTION (LOW): The skill has a surface for indirect prompt injection as it processes untrusted local data via Read/Grep. Evidence Chain: 1. Ingestion points: Local files accessed via Read/Grep tools. 2. Boundary markers: Absent. 3. Capability inventory: Read-only access (Read, Glob, Grep); no write or execute tools enabled. 4. Sanitization: Absent. The risk is considered minimal because the skill lacks the tools necessary to perform sensitive write operations or code execution.
Audit Metadata