secondme-openclaw-connect

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill asks the user to paste an authorization code (smc-...) into the chat and then explicitly instructs constructing a POST body containing that code and saving the returned accessToken into a credentials file, which requires the LLM to handle and embed secret token values verbatim (high exfiltration risk).

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly posts the user-provided auth code to the third-party API at https://app.mindos.com/gate/in/rest/third-party-agent/v1/auth/token/code, reads fields like accessToken, channelId, and relayUrl from that external response, and then updates local OpenClaw config—so untrusted third-party content can directly drive agent actions.