agent-vision-diagrams

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The skill instructs the agent to accept or obtain user tokens and then embed them verbatim in Authorization headers and CLI commands (e.g., "Authorization: Bearer "), which requires the LLM to handle secret values directly and risks exfiltration.

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

• Third-party content exposure detected (high risk: 0.85). The required workflow instructs the agent to fetch and read the hosted canonical skill doc from a user-provided app URL (<app-url>/skill.md), which is public web content authored by an outsider and becomes LLM-readable prose at runtime.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly instructs the agent at runtime to "fetch" and "follow" the hosted canonical skill doc at /skill.md, which would directly control agent instructions and is described as a required dependency for diagramming sessions.