skills/bmad-code-org/bmad-method-test-architecture-enterprise/bmad-testarch-test-review/Gen Agent Trust Hub
bmad-testarch-test-review
Fail
Audited by Gen Agent Trust Hub on Apr 26, 2026
Risk Level: HIGHEXTERNAL_DOWNLOADSPROMPT_INJECTIONCREDENTIALS_UNSAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The documentation in
resources/knowledge/playwright-cli.mdadvises users to globally install@playwright/cli@latest. This scope is not associated with the official Playwright project (Microsoft), presenting a high risk of typosquatting and the execution of unverified code if a malicious package is registered under that name. - [PROMPT_INJECTION]: The skill's primary workflow involves ingesting and analyzing untrusted test files from the project directory (
steps-c/step-02-discover-tests.md). This constitutes an indirect prompt injection surface (Category 8), as instructions embedded in the analyzed code could potentially influence the agent's logic or conclusions. - [CREDENTIALS_UNSAFE]: The knowledge base fragment
resources/knowledge/pact-broker-webhooks.mdrecommends creating GitHub Personal Access Tokens (PATs) with 'No expiration'. This promotes poor credential hygiene and increases security risks in the event of token compromise.
Recommendations
- AI detected serious security threats
Audit Metadata