contract-test

Pass

Audited by Gen Agent Trust Hub on Mar 23, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill includes instructions to operate in "AUTONOMOUS MODE" and explicitly tells the agent "Do NOT ask questions." These directives are intended to suppress interactive behavior and ensure the agent proceeds with the testing tasks without user interruption.
  • [EXTERNAL_DOWNLOADS]: The skill automates the installation of standard, well-known API testing and validation tools from official registries. This includes libraries like @stoplight/prism-cli, @stoplight/spectral-cli, and pact-foundation/pact for Node.js, and schemathesis and pact-python for Python. These actions are standard for the skill's stated purpose.
  • [COMMAND_EXECUTION]: The skill executes shell commands to perform package installations (npm install, pip install, gem install) and to run testing suites (npx prism, schemathesis run, npx vitest). These commands are necessary for the tool's functionality and target expected testing frameworks.
  • [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests and processes untrusted data from local project files (source code, route definitions, and API specifications).
  • Ingestion points: Scans files like package.json, requirements.txt, route registration logic, and openapi.yaml to build endpoint inventories.
  • Boundary markers: Absent. The skill does not wrap processed content in specific delimiters to prevent the agent from following instructions embedded within the code or specifications.
  • Capability inventory: The agent has the capability to execute shell commands (pip, npm, npx) and write telemetry data to the filesystem.
  • Sanitization: No explicit sanitization or validation of the content read from project files is performed before it is used to influence test generation.
  • [PERSISTENCE_MECHANISMS]: The skill attempts to write telemetry data to a hidden directory (~/.claude/projects/). This is used to record execution metadata for the /evolve pipeline, which is a standard feature for improving agent performance within the intended platform.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 23, 2026, 10:59 AM
Security Audit — agent-trust-hub — contract-test