spec-generator

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

• Third-party content exposure detected (high risk: 0.80). The workflow's Step 2c explicitly instructs the agent to "use web search" and Explore agents to investigate external codebases, APIs, or services, which requires fetching and interpreting open/public third-party content (docs, sites, forums) that can materially influence spec decisions and subsequent actions.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill calls notion-search against the data source collection://300c7bb8-aef1-8075-8429-000b28262f1f and uses notion-fetch to retrieve Notion page content at runtime, and that fetched content is injected into the agent's context to directly control spec-generation prompts.