pre-pr-review
Pass
Audited by Gen Agent Trust Hub on Nov 13, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes local Python scripts to manage the review workflow.
- It calls
shared/scripts/capture_check.pywith thestaleargument to verify if previous check results are still valid. - It utilizes
python3 shared/scripts/model_routing.pyto resolve model providers for independent reviews. - [INDIRECT_PROMPT_INJECTION]: The skill processes external data in the form of code diffs and requirement specifications, creating a potential surface for indirect prompt injection.
- Ingestion points: Code diffs and feature requirements processed during the review stages (SKILL.md sections 1 and 2).
- Boundary markers: Absent; the instructions do not specify the use of delimiters to separate untrusted code content from agent instructions.
- Capability inventory: The skill possesses the capability to execute local scripts and invoke other skills such as
full-review,verify-changes, andbrowser-smoke(SKILL.md section 2). - Sanitization: Absent; no explicit sanitization or escaping of the diff content is described before it is passed to the reviewer model.
Audit Metadata