agent-setup
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: 1.00). The SKILL.md explicitly instructs fetching a recipe from the external public URL https://fullstackrecipes.com/api/recipes/agent-setup, which the agent is expected to ingest to configure behavior, so untrusted third-party content could influence actions.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 0.90). The skill includes a runtime fetch (curl) from https://fullstackrecipes.com/api/recipes/agent-setup (recipe://fullstackrecipes.com/agent-setup) to obtain the recipe that directly controls agent configuration/instructions, so the external content is a required runtime dependency that can influence prompts.
Audit Metadata