test-analysis-extensions

Pass

Audited by Gen Agent Trust Hub on Aug 23, 2026

Risk Level: SAFE
Full Analysis
  • Test Reference Data: The skill is composed of static Markdown documents that map test framework features (assertions, markers, setup/teardown) for various languages. These files act as a structured knowledge base and do not perform any active operations or network requests.
  • Indirect Prompt Injection Mitigation: The skill includes explicit guidance for consuming agents to treat the provided files as data rather than authoritative instructions. This approach aligns with security best practices for handling external reference material during automated code analysis.
  • Capability Mapping: The capability tags (e.g., 'auto-edit', 'report-only') are used to inform the analyzing agent of what actions are safe to perform on a target codebase, enhancing the safety of the downstream analysis process.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 23, 2026, 12:34 AM
Security Audit — agent-trust-hub — test-analysis-extensions