Endform

dev.endformPublished by endform.devv0.82.2Updated Nov 4, 2026

Find failing, flaky, and slow Playwright tests on Endform and keep your suite healthy as it grows.

VerifiedSTDIODesktop onlyBrowser AutomationDeveloper Tools

Overview

AI-generated overview

Endform's local MCP server lets an assistant query Playwright test runs on Endform and debug failing, flaky or slow tests.

What you can do
  • Query test runs on Endform to find failing, flaky and slow Playwright tests
  • Investigate a specific failing test, such as one that only fails in CI
  • Debug a test step by step in a live test session
  • Fix or quarantine problematic tests so they stop blocking pull requests
  • Set up Endform so the suite runs there in CI
When to use it
Useful when a team already runs or wants to run its Playwright end-to-end suite on Endform and wants an assistant to triage failures, spot flaky or slow tests, and quarantine or fix them without leaving the chat.
Requirements
Runs as a local stdio process, typically launched with npx endform@latest mcp, so Node.js and network access are needed. The Claude Code, Codex and VS Code plugins add the server automatically. An Endform account and access to the project's test suite are implied; no environment variables or headers are declared.
Before you install
The CLI sends the Playwright tests you select, along with what they need to run and their results, to Endform, so review what is selected before running. The assistant can fix or quarantine tests, which changes your suite and CI behavior. Plugin updates are not automatic in Claude Code unless enabled.

Installation

In SourceWeft

  1. Open Endform in the dashboard and add it to a workspace.
  2. Enable the server for the chats that should use its tools.

Desktop only via STDIO. STDIO servers start a local process, so they need the SourceWeft desktop host.

Other MCP clients

Follow the launch instructions in the repository.

README

Endform skills

Endform runs your existing Playwright suite in parallel, up to 10x faster, and helps you keep it healthy as it grows.

The endform skill lets your agent run your tests on Endform and find failing, flaky, and slow tests. It can then fix or quarantine them, or debug a test step by step in a live test.

Install

Claude Code

bash
claude plugin marketplace add endformdev/skillsclaude plugin install endform@endformdev

Claude Code doesn't update plugins from this marketplace automatically. To turn that on, run /plugin, go to Marketplaces, select endformdev, and select Enable auto-update. To update once, run claude plugin update endform@endformdev.

Codex

bash
codex plugin marketplace add endformdev/skillscodex plugin add endform@endformdev

VS Code

In VS Code, enable chat.plugins.enabled, run Chat: Install Plugin From Source from the Command Palette, and enter https://github.com/endformdev/skills.

Other agents

Install the skill with skills.sh:

bash
npx skills add endformdev/skills --skill endform

Pick your agents when prompted, or pass them with -a, for example -a claude-code. Add -g to install it for all your projects.

Use

Ask your agent, for example:

  • "Set up Endform so our tests run on it in CI"
  • "Which of our e2e tests are flaky?"
  • "The checkout test only fails in CI, can you look into it?"
  • "Quarantine the flaky search test so it stops blocking our PRs"

You can also call the skill directly: /endform in Claude Code and Cursor, /endform:endform in VS Code, and $endform in Codex.

MCP server

The Endform MCP server lets your agent query your test runs and debug failing tests on Endform. The Claude Code, Codex, and VS Code plugins add it for you.

To add it without the plugin, in Claude Code:

bash
claude mcp add --transport stdio --scope local endform -- npx -y endform@latest mcp

For Cursor and other agents, see the MCP server docs.

Data and privacy

The plugin runs the Endform CLI from npm, which sends the Playwright tests you select, with what they need to run, and their results to Endform. See our privacy policy for more details.

Repository layout

skills/endform/SKILL.md      The skill: when to use Endform and which reference to readskills/endform/reference/    The reference the skill reads for each task: setup, CI, investigating tests, live testsplugin.json                  Plugin manifest for Codex, GitHub Copilot, VS Code, and Cursormcp.json                     MCP server for those agents.agents/plugins/             Marketplace for Codex.claude-plugin/              Plugin and marketplace for Claude Code.mcp.json                    MCP server for Claude Code

License

MIT

Source: README.md at commit a98ddb6

Tools

0
SourceWeft started this package once in an isolated sandbox, without credentials, to list its tools. That failed: the server did not complete the MCP handshake.

Version history

4
  1. v0.82.2LatestNov 4, 2026
  2. v0.82.1Nov 4, 2026
  3. v0.82.0Oct 29, 2026
  4. v0.81.5Oct 14, 2026

Score

50% of the applicable checks met

How complete this listing is, computed from the registry entry, the endpoint and public sources. Not a rating of the server and not a statement that it is safe or working today: reachability is a dated check, and no installs or reviews go into the score.

  • Reached in the last 30 days25 pts

    Keep the endpoint answering the MCP handshake; SourceWeft probes it every 30 days.

  • Auth declared10 pts

    Declare the credential headers in the registry entry, or keep the endpoint answering the probe so its auth can be classified.

  • Tools listed15 pts

    Answer tools/list without a credential, or ship a package that starts in a sandbox without one.

  • DescriptionRequired15 pts
  • Logo10 pts
  • Documentation15 pts
  • License10 pts