remember

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt instructs the agent to accept arbitrary user-provided text, pass it verbatim into an add_memory API call, and echo the stored content (first 80 chars) back to the user—so any secrets (API keys, tokens, passwords) included would be sent in requests and returned in output.