finding-triage

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

• Third-party content exposure detected (high risk: 0.75). The required runtime workflow fetches Hacktron findings and finding details via curl from https://api.hacktron.ai/v1/findings... and .../v1/findings/<id>, where the description, proof_of_concept, and triage_thread fields can contain outsider-authored free text (e.g., comments/threads) that the agent then ingests into its LLM context.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill explicitly requires fetching and reading the live docs at runtime (e.g., https://docs.hacktron.ai/llms.txt and the linked .md pages) and instructs the assistant to follow those documents to determine API endpoints/fields and construct requests, so the externally fetched content directly controls the agent's instructions and is a required runtime dependency.