pw
Pass
Audited by Gen Agent Trust Hub on Nov 6, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user-supplied data (such as user stories, application URLs, or existing test code) to generate, review, or fix Playwright test scripts.
- Ingestion points: External content enters the agent's context through parameters in commands like
/pw:generate <spec>,/pw:fix <test>, and/pw:migrate(SKILL.md). - Boundary markers: The skill does not implement explicit delimiters or instructional warnings to prevent the agent from obeying malicious commands embedded within the input specifications.
- Capability inventory: The skill is designed to write executable TypeScript/JavaScript files to the filesystem and read existing project code. It also encourages running these tests, which may involve network operations (SKILL.md).
- Sanitization: There are no documented mechanisms for validating or sanitizing the input data before it is interpolated into the code generation prompts.
Audit Metadata