a2ui-issue-triage

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

• Third-party content exposure detected (medium risk: 0.30). In the required runtime workflow, the agent executes scripts/fetch_issues.py to read outsider-authored GitHub issue titles/bodies and comments from the A2UI repository (gh issue list then gh issue view ... --json comments), and then passes that free text into the analysis/suggestion steps.

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

• Potentially malicious external URL detected (high risk: 0.90). The dashboard HTML (assets/triage_dashboard.html) loads and executes remote JavaScript at runtime from the CDN URLs https://cdn.jsdelivr.net/npm/marked/marked.min.js and https://cdn.jsdelivr.net/npm/dompurify/dist/purify.min.js, which the local dashboard depends on for markdown parsing/sanitization and therefore are required runtime-executed external dependencies.