testing-review-contract

Define AgentHub v1.0+ testing layers, contract validation, and CI quality gates.

Updated May 19, 2026
One-click install
npx skills add https://github.com/zxc1a1a1/Multi_Agent-AgentHub --skill testing-review-contract
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: testing-review-contract
Source: https://github.com/zxc1a1a1/Multi_Agent-AgentHub/tree/main/.agents/skills/testing-review-contract
Command: npx skills add https://github.com/zxc1a1a1/Multi_Agent-AgentHub --skill testing-review-contract

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill prevents fragile or unsafe changes in AgentHub by enforcing a contract-first, security-aware test and review standard across services and runtime behaviors.

Core Features & Use Cases

  • Contract-first and schema validation: Ensures every cross-boundary interface has valid/invalid fixtures and fails safely before real execution.
  • Deterministic CI testing layers: Defines a test matrix covering unit, component/store, adapter/converter, service integration, and cross-process integration (Gateway ↔ Orchestrator).
  • Failure, streaming, artifacts, and security guardrails: Requires replayable streaming/SSE tests, artifact-to-runtime capability validation, fallback/retry correctness, and security regression checks (redaction, iframe sandboxing, confirm_action for high-risk tools).
  • PR Review checklist and CI quality gates: Makes merge contingent on passing the required gates and verifying that contracts, fixtures, and process separation remain intact.

Quick Start

Use the testing-review-contract Skill as your PR and CI checklist by updating/adding schema/fixture coverage for every affected contract and running the required unit/contract/security checks before requesting review.

Frequently Asked Questions about testing-review-contract

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I enforce contract-first testing for multi-agent orchestration services?

Contract-first testing enforces strict schema validation and deterministic fixtures across multi-agent boundaries. This standard ensures interfaces fail safely using valid and invalid inputs before real execution, blocking unsafe merges across cross-process services like Gateway and Orchestrator.

What is the best way to test SSE streaming events in a CI pipeline?

Testing SSE streaming events requires replayable streaming tests within your CI pipeline. This approach validates streaming behaviors deterministically, ensuring that cross-protocol changes and artifact-to-runtime capabilities are verified before integration.

How do I set up CI quality gates to block unsafe merges across microservices?

CI quality gates block unsafe merges by making PR reviews contingent on passing defined test matrices. You set up gates covering unit, component, and cross-process integration tests, ensuring contracts, fixtures, and process separation remain intact.

Does contract-first schema validation work with Docker demo paths and persistence layers?

Contract-first schema validation applies to Docker demo paths and persistence layers. It requires deterministic fake inputs and strict fixture rules to validate cross-boundary interfaces, ensuring changes across these runtime behaviors fail safely.

Why do I need security regression checks for high-risk agent tools?

Security regression checks are needed to prevent unsafe changes by validating redaction, iframe sandboxing, and confirm_action for high-risk tools. These controls ensure security guardrails remain intact during cross-service and cross-protocol modifications.

What are the limitations of relying on unit tests for multi-agent orchestration changes?

Unit tests alone cannot verify cross-process integration or streaming behaviors. Limitations include missing failure-path coverage, fallback correctness, and security regression controls, requiring a comprehensive test matrix to prevent fragile multi-agent changes.