zudoesa-articlify

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The prompt explicitly instructs preserving the conversation verbatim in --conversation mode, which would force the agent to reproduce any API keys, tokens, passwords, or other secrets present in the conversation exactly, creating a direct exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.70). The required workflow delegates to zudoesa-writer using a prompt that (in --conversation mode) must preserve the actual conversation verbatim; that conversation content is outsider-authored free text when it includes messages/comments from other participants, which the subagent then ingests into the LLM context via the constructed writing brief.