autonomous-tests

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The skill explicitly requires showing the full config to the user for trust approval and requires including assigned credentials in agent task descriptions — actions that will force verbatim exposure of any raw secret values present in the config or userContext (even though it recommends using env-var names, it does not prevent raw secrets from being output).