mediamtx-docs

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

• Third-party content exposure detected (high risk: 0.75). The skill fetches and text-extracts public web content from https://mediamtx.org/docs/<page> at runtime (fetch_url() → extract_text() → printed/returned text), which is then placed into the agent’s LLM context via the skill’s fetch/search/section outputs (indirect prompt-injection risk from untrusted page text).