reflection-method-call

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill's examples show embedding full JSON (including serialized "value" fields) directly into a CLI --input argument, so any secrets placed in those fields would be emitted verbatim in generated commands/outputs (exfiltration risk).

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This skill exposes an ability to invoke any C# method (including private methods) and to deserialize arbitrary objects to act as call targets/parameters, which directly enables remote code execution, credential/data exfiltration, and other backdoor behaviors if misused.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). This skill’s required runtime workflow is “reflection-method-call,” which takes an outsider-provided JSON payload (e.g., filter, targetObject, inputParameters) and uses it to reflectively call C# methods in the project, potentially ingesting free text that is used to select/deserialize parameters (via filter/method schema) before any specific trusted item is selected.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (high risk: 1.00). This skill explicitly enables invoking any C# method (including private methods) with arbitrary deserialized inputs and new-instance creation, which permits calling system-modifying APIs (file operations, process execution, user management, etc.) and thus can compromise the host machine state.