tapestry
Warn
Audited by Snyk on Mar 1, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 1.00). SKILL.md explicitly fetches and processes arbitrary external URLs in Step 2 (YouTube transcripts via yt-dlp, articles via curl/trafilatura/reader, and PDFs via curl/pdftotext) and then automatically generates action plans from that extracted content, so untrusted third‑party pages can directly influence the agent's decisions and tool use.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The skill fetches and reads arbitrary user-provided URLs at runtime (e.g., https://www.youtube.com/watch?v=dQw4w9WgXcQ and https://example.com/research-paper.pdf) and then directly uses that extracted content to generate the Ship-Learn-Next action plan, meaning remotely hosted content can directly control the agent's prompts/outputs.
Audit Metadata