test-specialist

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references industry-standard testing tools and services in its documentation and examples, including Vitest, Jest, Playwright, Supertest, Percy, and Chromatic. These are well-known services commonly used in software development workflows.
  • [INDIRECT_PROMPT_INJECTION]: The skill contains Python scripts (analyze_coverage.py, find_untested_code.py) that ingest and process external data, such as test coverage reports and directory structures. This represents a potential surface for indirect prompt injection, although the scripts themselves are limited to parsing and reporting metrics.
  • Ingestion points: scripts/analyze_coverage.py (reads JSON coverage files); scripts/find_untested_code.py (scans directory structures).
  • Boundary markers: None present in the utility scripts.
  • Capability inventory: The scripts perform local file system analysis and JSON parsing, reporting results back to the agent.
  • Sanitization: The scripts rely on standard library parsing (json, pathlib) and do not implement specific content sanitization for the data contained within the reports.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 08:36 PM
Security Audit — agent-trust-hub — test-specialist