payload

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs users to provide credentials (email/password), extract and reuse session cookies/payload-token, and includes example curl commands that embed those secrets in requests, which requires the LLM to handle and potentially emit secret values verbatim.

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

• Third-party content exposure detected (high risk: 0.80). The skill's required workflow (SKILL.md, "Step 1: Determine the API Endpoint" and the "Verify" step) explicitly instructs the agent to fetch and query arbitrary user-provided/public Payload site URLs (e.g., curl https://your-site.com/api/posts...), and the agent is expected to read and act on those API responses (existence checks, create/update decisions), exposing it to untrusted third-party CMS content that could carry indirect prompt-injection content.