MCPInstall

integration-tests

by openai OfficialTesting & QA

Run the integration-tests pipeline that depends on a local npm registry (Verdaccio). Use when asked to execute integration tests or local publish workflows in this repo.

Installs
110
Last 24h
—
Trending rank
—
Security
Not audited

Install

Works with every agent the skills CLI supports.

npx skills add https://github.com/openai/openai-agents-js --skill integration-tests
  1. 01

    Install

    Run the command in your project. The CLI asks which agents to install for.

  2. 02

    Describe the task

    Ask your agent for something this skill covers. It loads integration-tests automatically when relevant.

  3. 03

    Review the result

    The agent follows the SKILL.md steps, scripts and references bundled with the skill.

SKILL.md

The instructions your agent reads when the skill is triggered.

Integration Tests

Overview

Run integration tests that require a local npm registry by starting pnpm local-npm:start in a background subprocess, executing the build/reset/publish/test pipeline in the main process, then cleanly shutting down the registry process and returning results.

Workflow

Codex execution environment

  • When Codex invokes this skill, run both pnpm local-npm:start and the main pipeline outside the Codex sandbox by default (sandbox_permissions=require_escalated). The integration suites install dependencies inside fixture projects with npm install, bun install, Deno, Wrangler, local emulators, Docker, and other subprocesses; running them inside the Codex sandbox can cause environment-only failures such as npm EPERM writing to ~/.npm/_cacache/tmp or Bun PermissionDenied writing to its tempdir.
  • Use sandboxed execution only when the user explicitly asks for it. If a sandboxed run fails during dependency installation or temp/cache writes, rerun the same exact pipeline outside the sandbox before classifying the failure as repo-induced.
  • Still keep the registry lifecycle local to the run: start Verdaccio before the pipeline and stop it after the pipeline, even when a failure occurs.

1. Start the local registry (subprocess)

  • Start a background process with pnpm local-npm:start and keep its session id so it can be stopped later.
  • Wait until the registry is ready (look for a Verdaccio listen message or the default http://localhost:4873 line). If no explicit ready line appears, wait a few seconds and proceed.
  • If the port is already in use, note that an existing registry may be running and proceed only if it matches the expected local registry; otherwise stop it and restart.

2. Run the main pipeline (main process)

Run this exact sequence in the main process and capture the output:

pnpm i && pnpm build:ci && pnpm local-npm:reset && pnpm local-npm:publish && OPENAI_AGENTS_RUN_STORAGE_MOUNT_INTEGRATION=1 pnpm test:integration
  • Use pnpm build:ci here so the skill validates the same serialized build path that GitHub Actions now uses, while still running the normal prebuild and postbuild lifecycle steps.

  • Run pnpm test:integration with OPENAI_AGENTS_RUN_STORAGE_MOUNT_INTEGRATION=1 so the optional sandbox storage mount coverage runs by default when this skill is used. This enables the local Azurite and MinIO smoke tests without changing the repository's plain pnpm test:integration default.

  • Return the full success/failure outcome and a concise summary of the results.

  • Always capture the stdout/stderr from pnpm test:integration and include it in the final response (trim obvious noise if extremely long) inside a fenced code block.

  • Do not use watch mode.

If pnpm local-npm:publish fails

Troubleshoot using integration-tests/README.md, which lists the canonical recovery steps. If time is short, prioritize the fixes in the order given there and surface the exact error text in your response.

3. Clean up the registry process

  • Send Ctrl+C to the registry subprocess and wait for it to exit.
  • If it does not exit, terminate it by PID and confirm the port is free before finishing.

Output expectations

  • Always include the integration test results in the response.
  • If any step fails, include the failing command, the error output summary, and the next recommended action.

FAQ

What does the integration-tests skill do?

Run the integration-tests pipeline that depends on a local npm registry (Verdaccio). Use when asked to execute integration tests or local publish workflows in this repo.

How do I install integration-tests?

Run npx skills add https://github.com/openai/openai-agents-js --skill integration-tests in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/integration-tests/ (personal) or .claude/skills/integration-tests/ (project).

Which AI agents can use integration-tests?

Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.

Is integration-tests safe to install?

No partner security audit is available yet. Skills can run commands and scripts, so read the SKILL.md and any bundled files before installing.

Who maintains integration-tests?

It is published by openai in openai/openai-agents-js, and it is an official skill from the maker of the technology it covers. It has 110 recorded installs.

More from openai/openai-agents-js

View all →

docs-sync

openai/openai-agents-js

Audit or update authored English SDK documentation within a requested topic or diff using implementation evidence.

OfficialDocs & Writing147 installs

pnpm-upgrade

openai/openai-agents-js

Open the skill to read its SKILL.md instructions.

OfficialAI & Agents143 installs

code-change-verification

openai/openai-agents-js

Run the required final install, build, type, lint, test, and format checks after eligible SDK changes pass review.

OfficialAI & Agents139 installs

test-coverage-improver

openai/openai-agents-js

Open the skill to read its SKILL.md instructions.

OfficialTesting & QA132 installs

openai-knowledge

openai/openai-agents-js

Retrieve authoritative OpenAI API and platform documentation when an integration or claim needs current external evidence.

OfficialAI & Agents130 installs

pr-draft-summary

openai/openai-agents-js

Prepare a local PR title, description, and branch suggestion when requested or after eligible implementation work.

OfficialCode Quality & Workflow129 installs

Popular testing & qa skills

View category →

tdd

mattpocock/skills

Test-driven development. Use when the user wants to build features or fix bugs test-first, mentions "red-green-refactor", or wants integration tests.

Testing & QA1.2M installs

grilling

mattpocock/skills

Grill the user relentlessly about a plan, decision, or idea. Use when the user wants to stress-test their thinking, or uses any 'grill' trigger phrases.

Testing & QA996.2K installs

migrate-to-shoehorn

mattpocock/skills

Migrate test files from as type assertions to @total-typescript/shoehorn. Use when user mentions shoehorn, wants to replace as in tests, or needs partial test data.

Testing & QA484.1K installs

run-train

lllllllama/rigorpilot-skills

Rigor Train skill for deep learning research repositories. Use when a documented or selected training command should be run conservatively for startup verification, short-run verification, full kickoff, or resume, with command, config, seed, log, checkpoint, status, and metric evidence written to standardized trainoutputs/. Do not use for environment setup, exploratory sweeps, speculative idea implementation, or end-to-end orchestration.

Testing & QA337.8K installs

test-driven-development

obra/superpowers

Use when implementing any feature or bugfix, before writing implementation code

Testing & QA261.8K installs

Use when implementation is complete, all tests pass, and you need to decide how to integrate the work

Testing & QA216.8K installs