entry-point-analyzer

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

• Third-party content exposure detected (high risk: 0.70). The required workflow ingests the analyzed smart contract source files (step 3–4: “Locate Contracts” and “Extract Entry Points”), which are outsider-authored if the operating user points the skill at a third-party codebase, and those files’ readable code text is then parsed into the agent’s LLM context for report generation.