muapi-workflow

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt's CRITICAL RULE forces the agent to ask for and then use exact "API inputs" provided by the user before executing workflows, which requires the LLM to receive and embed potentially secret API keys or passwords verbatim in commands/tool calls, creating an exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly instructs the agent to fetch and parse external workflow catalogs and definitions via the muapi CLI (e.g., muapi workflow discover and muapi workflow get — including the "API Inputs" table), which are third-party/user-provided workflow descriptions the agent must read and which can materially influence workflow selection and execution.