review

Pass

Audited by Gen Agent Trust Hub on Mar 8, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill contains a vulnerability surface for indirect prompt injection.
  • Ingestion points: The skill ingests untrusted data from the 'changedFiles' and 'proof' fields as defined in the Input table of references/flow.md.
  • Boundary markers: The prompt template used for the '/ask' command lacks delimiters (such as triple backticks or XML tags) to isolate the untrusted input data from the instructions.
  • Capability inventory: The skill utilizes the '/file-op' (read_file) capability to access local repository content and the '/ask' tool to communicate with external AI providers.
  • Sanitization: There is no evidence of input validation, sanitization, or escaping of external content before it is interpolated into the prompts for the secondary models.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 8, 2026, 06:12 AM