agent-ingestor

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt requires the agent to read external prompt/config folders and "Show the Source Code" / output the exact text it will write (including contents of provided files), which would force verbatim reproduction of any secrets or API keys present in those inputs.

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

• Third-party content exposure detected (high risk: 0.75). The skill’s required runtime workflow is an ingestion protocol that operates on user-provided prompt/configuration inputs (and potentially “external configuration folder” contents), which are outsider-authored free text that the agent would read and then inject into LLM context during analysis/deconstruction.