best-practices-audit

Pass

Audited by Gen Agent Trust Hub on Mar 7, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted content from the project's CLAUDE.md file, which could contain malicious instructions designed to influence the agent's behavior during the audit or ship phase.\n
  • Ingestion points: Reads the project root CLAUDE.md and associated documentation files in the docs/ directory.\n
  • Boundary markers: No specific delimiters or boundary markers are defined to isolate untrusted content in the prompt context.\n
  • Capability inventory: The skill can read and write files, restructure documentation, and execute shell commands to open HTML reports.\n
  • Sanitization: The skill implements a strict sanitization process for dynamic parameters used in shell commands, including a character whitelist and the use of end-of-options markers to prevent flag injection.\n- [COMMAND_EXECUTION]: The skill executes system commands (open or xdg-open) to display visual audit reports to the user. This behavior is mediated by a multi-step sanitization logic that checks for shell metacharacters and enforces a strict naming schema for the file path.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 7, 2026, 06:34 AM