testing-handbook-generator

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). Yes—during runtime the skill’s required workflow fetches and ingests non-video external documentation/blog content from URLs listed in the handbook’s 99-resources.md/91-resources.md via WebFetch, which is outsider-authored free text that can include prompt-injection payloads.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The generator explicitly fetches the Testing Handbook at runtime (e.g., https://github.com/trailofbits/testing-handbook and git clone https://github.com/trailofbits/testing-handbook.git), and that fetched repository content is required and directly drives the agent prompts/instructions used to generate skills.