framing-doc

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs the agent to read user-specified transcript files and "capture verbatim quotes" into generated output, which would reproduce any secrets (API keys, tokens, passwords) present in those transcripts verbatim.

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

• Third-party content exposure detected (medium risk: 0.30). SKILL.md describes a workflow that reads “one or more conversation transcripts” provided by the user via specific file paths, and then captures verbatim quotes from those outsider/authored transcripts—thus ingesting arbitrary free text at runtime through user-supplied transcript contents.