Connects supported AI clients, including Claude Code and Codex, to OpenWork through MCP, a protocol for accessing external tools. It guides account sign-in and verifies that tools for finding and running organization capabilities are available.
Connect the OpenWork MCP Gateway (https://api.openworklabs.com/mcp/agent) to Claude Code, Codex, Gemini CLI, Cursor, VS Code, Claude Desktop, or ChatGPT so the agent can use the user's OpenWork organization skills, plugins, and connections.
Review the instructions the agent will receive before you install.
Server URL: https://api.openworklabs.com/mcp/agent
Transport: Streamable HTTP. Auth: OAuth (PKCE, dynamic client registration); the client opens a browser for sign-in. No API key is needed.
The user needs an OpenWork Cloud account. If they don't have one, send them to https://app.openworklabs.com?mode=sign-up first.
Use the command for the client you are running in. If an openwork entry already exists, don't add a duplicate; sign in with the existing one.
Claude Code:
claude mcp add --transport http openwork https://api.openworklabs.com/mcp/agent
# add `-s user` to make it available in every project
Codex:
codex mcp add openwork --url https://api.openworklabs.com/mcp/agent
codex mcp login openwork
Gemini CLI:
gemini mcp add --transport http openwork https://api.openworklabs.com/mcp/agent
OpenCode:
opencode mcp add openwork --url https://api.openworklabs.com/mcp/agent
opencode mcp auth openwork
Cursor, VS Code, Claude Desktop, ChatGPT, Windsurf, Zed: add https://api.openworklabs.com/mcp/agent as a remote MCP server. Per-client steps: https://openworklabs.com/docs/start-here/connect-openwork-mcp
/mcp, select openwork, and authenticate.codex mcp login openwork opens the browser./mcp auth openwork.opencode mcp auth openwork opens the browser.The user signs in and picks their organization. The organization is pinned to the token; to switch, log out of the openwork server and sign in again.
claude mcp list (or the client's equivalent) shows openwork.search_capabilities and execute_capability tools.Don't claim the connection works until step 3 passes.
invalid_grant: log out of openwork in the client and sign in again.https://api.<your-den-web-host>/mcp/agent.No reviews yet. Tell others what you thought.
No login needed. Share how it worked for you, ask a question, or note where it did not run.
No comments yet. Be the first to write one.
Manage OpenWork feature releases with feature flags, switches that control access to new behavior. Register features before coding, release them by organization and deployment, prepare a global shutdown option, and remove temporary flags after rollout.
Connect OpenCode browser tools to a locally running OpenWork Electron app to inspect its UI, enter a task, and check the response. This skill supports exploration and debugging through CDP, Chrome's remote control protocol.
Check added lines in a change for information identifying customers, prospects, partners, or outside individuals. Report locations and generic remediation without reproducing identities, and escalate cases where the information is already public.
Review an outside contributor’s pull request before running its code. Identify concealed behavior, obfuscated code, risky dependencies, and build hooks, with file locations and specific checks for maintainers to follow up.
Create an OpenCode plugin for OpenWork with custom tools and hooks, which run code at specific events. The skill builds the plugin file and adds its entry to the project configuration.
Set up and troubleshoot Daytona sandbox environments, which provide isolated spaces for running applications. The skill covers CLI authentication, persistent desktop and server sandboxes, diagnostic logs, shared secrets storage, snapshot refreshes, and environment deletion.