flux-opencode-remember

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly extracts and forwards arbitrary user content (e.g., "the API key is in Vault") to another skill for persistent writing, which means any secrets provided by the user would be included verbatim in the agent's output/delegation and could be exfiltrated.