failure-injection-test-author

Pass

Audited by Gen Agent Trust Hub on Aug 12, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references official Docker images from well-known sources, specifically wiremock/wiremock:3 and ghcr.io/shopify/toxiproxy:latest.
  • [EXTERNAL_DOWNLOADS]: Employs standard Node.js packages such as axios, jest, and toxiproxy-node-client for test execution and HTTP communication.
  • [COMMAND_EXECUTION]: Uses common development tools including docker compose, curl, and npx to set up infrastructure and run test suites.
  • [INDIRECT_PROMPT_INJECTION]: Ingests tool outputs such as WireMock request logs and test results. This represents a standard testing ingestion surface. Evidence: 1. Ingestion points: /__admin/requests in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: npx jest, docker compose, curl. 4. Sanitization: Absent. This follows standard testing practices for verifying system behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 12, 2026, 05:02 PM
Security Audit — agent-trust-hub — failure-injection-test-author