gherkin-bdd-reviewer
Pass
Audited by Gen Agent Trust Hub on Aug 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-provided BDD files as part of its review workflow.
- Ingestion points: The agent reads .feature files and source code bindings (e.g., .cs, .java, .js) from the user's repository during artifact location and review.
- Boundary markers: The skill provides a structured
review-checklist.mdto define analysis boundaries, although it lacks specific delimiters for raw file content. - Capability inventory: The skill possesses the capability to read local files, write reports to the
docs/reviews/directory, and execute local bash and python3 commands for metadata processing. - Sanitization: The
collect-diff.shscript uses python3 to JSON-escape the diff payload, preventing content from breaking the JSON structure used to communicate with the agent. - [COMMAND_EXECUTION]: The skill executes local bash scripts to identify artifacts and framework types. These scripts use standard tools such as
git,find, andgrepfor static analysis and do not incorporate unvalidated external input into shell commands in a way that allows for privilege escalation.
Audit Metadata