ai-orchestrator
Warn
Audited by Snyk on Mar 2, 2026
Risk Level: MEDIUM
Full Analysis
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.90). This skill explicitly dispatches web-research/web-fetch tasks and instructs agents to fetch arbitrary upstream documentation and web pages (see examples/dispatch-web-fetch.md, SKILL.md "Upstream docs — fetch on demand" and references/sources.md) so the spawned agents will read untrusted public URLs (WebFetch/Gemini/CLI) and act on that content.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.80). The skill includes an explicit runtime example that instructs a spawned agent to fetch and ingest external content (which directly becomes the agent's context) from URLs such as https://example.com/blog/interesting-post via the web-fetch skill, so this is a runtime external dependency that can control prompts.
Audit Metadata