audit-stale-comments

Pass

Audited by Gen Agent Trust Hub on Aug 4, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes git ls-files and rg (ripgrep) to discover source code files within the repository. These are standard, benign utilities for file discovery in a development context.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes untrusted text from source code comments which could contain adversarial instructions targeting the agent. \n- Ingestion points: Source files (JavaScript, TypeScript, JSX, TSX, Go) provided as arguments or discovered via git/rg. \n- Boundary markers: Absent; the instructions do not specify delimiters to separate untrusted code content from the agent's internal logic. \n- Capability inventory: File discovery via shell commands and read-only file access. The skill does not have network access or write permissions by default. \n- Sanitization: None; the skill does not filter or escape comment content before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 4, 2026, 08:03 AM
Security Audit — agent-trust-hub — audit-stale-comments