dream

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill instructs the agent to grep and extract content from raw session JSONL transcripts and then write concise memory entries into persistent topic files (MEMORY.md and topic files) without any guidance to redact or avoid secrets, so any API keys, tokens, or passwords present in sessions could be copied verbatim into memory and outputs, causing exfiltration risk.