search-past-chats

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill instructs the agent to retrieve and display full past session messages and excerpts (via get-session and search results), which can include verbatim API keys, tokens, or passwords from prior chats, so the LLM would likely output secret values directly if they exist in history.