one-step-better-ai-pm

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

• Third-party content exposure detected (high risk: 0.90). The skill fetches and parses latest briefs from the public API (WebFetch https://genaipm.com/api/feed/latest?email=) and in Phase 4 explicitly extracts URLs from brief HTML and uses WebFetch to read original articles/blogs/docs on the open web, which the agent then reads and uses to decide and apply code changes—exposing it to untrusted third‑party content that could contain indirect prompt injections.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill performs a runtime WebFetch of the GenAI PM feed at https://genaipm.com/api/feed/latest?email= (and then extracts and fetches URLs from the brief HTML) and uses that fetched content to drive the agent's decision-making and apply improvements, so external content directly controls the agent's instructions.