mem-setup

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill requires the agent to accept a user-provided secret (the cm_ sync token) and write it verbatim into plugin config files (via Write/Edit), which forces the LLM to handle/output the exact secret despite instructions not to echo it in conversation.

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

• Third-party content exposure detected (low risk: 0.10). This skill only ingests user-pasted cmem.ai Connect values in the chat flow to extract sync token/user id/syncHubUrl, and it does not describe reading any outsider-authored third-party text from feeds, queues, or repositories.