notion-api

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly instructs asking for and using a user-provided API key ("use that instead") and shows Authorization headers for curl, which can lead the LLM to embed the secret verbatim in generated requests/commands (exfiltration risk) despite recommending env vars.

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

• Third-party content exposure detected (medium risk: 0.65). The skill’s runtime flow can call Notion endpoints like “Retrieve Block Children” (GET /v1/blocks/{block_id}/children) and “Retrieve Comments” (GET /v1/comments?...), which return outsider-authored page/block/comment text that may be inserted into the agent’s LLM context (indirect prompt injection risk).