fail-fast
Pass
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted source code from external repositories provided by the user. Malicious instructions embedded in the target code or comments could potentially influence the agent's behavior during the refactoring process.\n
- Ingestion points: The
scripts/scan-fallbacks.pyscript reads the contents of the target repository's files.\n - Boundary markers: There are no explicit instructions or delimiters to isolate or warn the agent about potentially malicious instructions within the codebase findings.\n
- Capability inventory: The agent is granted permission to read files via the scanner and perform narrow edits to the codebase, including adding or updating tests.\n
- Sanitization: The skill lacks sanitization or validation of the text findings returned by the scanner before the agent processes them.\n- [COMMAND_EXECUTION]: The skill requires the execution of a Python scanner script (
scripts/scan-fallbacks.py) and utilizes project-local tools like@howells/envyvianpx. These operations are performed locally on the user-provided codebase and are consistent with the skill's purpose.
Audit Metadata