soultrace

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

• Third-party content exposure detected (high risk: 0.80). The SKILL.md Agent Interaction Protocol requires calling the public third-party endpoint https://soultrace.app/api/agent and presenting/acting on the question text returned by that API (each response's "question.text"), so the agent ingests and uses external, untrusted web content as part of its workflow.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill makes runtime calls to https://soultrace.app/api/agent to fetch each question (the prompt text) which directly controls what the agent presents to the user and is required for the test.