recipe-review

Pass

Audited by Gen Agent Trust Hub on Mar 13, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface.
  • Ingestion points: Processes data from project design documents (docs/design/*.md) and output from git diff commands.
  • Boundary markers: The orchestrator does not use explicit delimiters or specific 'ignore instructions' warnings when passing external document content to sub-agents.
  • Capability inventory: The workflow can invoke a task-executor sub-agent that has permissions to modify local files and apply code changes.
  • Sanitization: Input content from design documents is not sanitized or filtered before being interpolated into sub-agent prompts.
  • [COMMAND_EXECUTION]: Executes standard shell commands (ls, grep, git diff) to identify documentation and implementation files. These operations are local, restricted to the project directory, and serve the primary function of quality assurance validation.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 13, 2026, 11:03 AM