api-integration

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

• Third-party content exposure detected (high risk: 0.75). SKILL.md:16-49 requires running devtools_discovery → search_docs searches, which at runtime can pull in outsider-authored public documentation text (web pages/docs) and then the workflow explicitly synthesizes/extracts that content into the agent’s LLM context (step 4: SKILL.md:57-58).