qa-api-testing-contracts

Pass

Audited by Gen Agent Trust Hub on Oct 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted API artifacts, creating an indirect prompt injection surface. * Ingestion points: API artifacts are collected as inputs and processed by scripts/schemathesis_baseline.sh and scripts/spectral_lint.sh. * Boundary markers: Not present in SKILL.md. * Capability inventory: Command execution and network operations in scripts/pact_can_i_deploy.sh and scripts/schemathesis_baseline.sh. * Sanitization: Not present in the provided bash scripts.
  • [COMMAND_EXECUTION]: The bash scripts in the scripts/ directory execute several CLI tools: pact-broker, schemathesis, buf, and spectral.
  • [EXTERNAL_DOWNLOADS]: The scripts/schemathesis_baseline.sh script uses uvx to download and run the schemathesis package from the Python Package Index (PyPI), which is a well-known service.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 6, 2026, 02:03 PM