shiplog

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly requires constructing curl requests with an Authorization header containing the XAI_API_KEY (Authorization: Bearer {XAI_API_KEY}), which forces the agent to include the secret verbatim in generated commands/requests.

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

• Third-party content exposure detected (medium risk: 0.30). In Step 3, the workflow builds a free-form prompt containing outsider-authored X post text (searched from operator/product/scout accounts) and ingests the resulting tweet content from the X.AI https://api.x.ai/v1/responses output, and Step 3/4 also optionally WebFetch https://x.com/<handle> pages as a lower-quality fallback.