iot

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

• Third-party content exposure detected (medium risk: 0.65). The required runtime workflow uses the MCP tools to “search/read/recommend documentation” which can fetch and ingest publicly authored web documentation text into the agent’s LLM context, creating an indirect prompt-injection path.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill contains installation and Greengrass component artifact URIs that fetch and execute remote code at install/runtime (curl to https://d2s8p88vqu9w66.cloudfront.net/releases/greengrass-nucleus-latest.zip and component artifacts at s3://acme-greengrass-artifacts/... ), so these external URLs are runtime dependencies that can execute remote code.