claim-chart

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt requires verbatim quoting of local files, source code, and produced documents into markdown/CSV/spreadsheets (and to write those files), so if those sources contain secrets (API keys, tokens, passwords, cookies) the LLM is explicitly instructed to emit them verbatim, creating an exfiltration risk.