setting-up-a-custom-rest-source

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill explicitly instructs the agent to ask the user for credential values and include them verbatim in tool/request payload fields like auth_token/auth_api_key/auth_password, meaning the LLM must handle and emit secrets directly.

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md instructs fetching user-provided API “docs URL” at Step 1 and using that content to draft the manifest, which implies runtime ingestion of outsider-authored free text from arbitrary external web pages into the LLM context.