specifying-gates

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill asks users to supply exact CLI commands, HTTP requests, or subagent briefings and then writes those strings verbatim into task metadata and descriptions (e.g., as verifyCommand or subagentBrief), so if those strings contain API keys, tokens, or passwords the LLM will handle and output secret values directly.