review-spec-local
Pass
Audited by Gen Agent Trust Hub on Nov 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes multiple Python scripts located in the .github/scripts/ directory of the repository (e.g., prepare_local_review_inputs.py, validate_review_json.py) to manage the review workflow.
- [DYNAMIC_EXECUTION]: The workflow dynamically determines a secondary review skill to execute based on output from a local script, leading the agent to follow instructions from a path computed at runtime.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from pull request diffs and descriptions which could contain malicious instructions.
- Ingestion points: pr_description.txt and pr_diff.txt.
- Boundary markers: Absent; the instructions do not specify delimiters or system prompts to ignore embedded instructions in the ingested data.
- Capability inventory: The skill possesses file system read/write permissions and the ability to execute local scripts via python3.
- Sanitization: Absent; there is no process mentioned for sanitizing the pull request content before it is provided to the agent.
Audit Metadata