MCP server · General-purpose
A virtual computer your agent spins up for any task.
OpenHelm gives your assistant a secure remote environment: a stealth browser, a code runner, file access and more. Write a task in natural language, get back structured, evidence-backed results. Zero-install MCP server or REST API, your choice.
What it does
The general-purpose server gives your assistant the keys to a full virtual computer. Rather than a fixed opinionated workflow, it exposes the OpenHelm /v1 engine directly: define a task in natural language, optionally pin a JSON output schema, and the agent executes it inside an isolated cloud environment with a stealth browser and code execution, returning structured, evidence-backed results.
It is the right server when you are building custom automations, prototyping in Cursor, or want full control over prompts and output shape rather than a domain-specific toolset.
Endpoint
Add this URL to your MCP client.
https://mcp.openhelm.ai/main/mcpnpx mcp-remote https://mcp.openhelm.ai/main/mcpTools
What your assistant can call.
Available on https://mcp.openhelm.ai/main/mcp
run_taskExecute an inline natural-language prompt, with an optional structured JSON output schema.
create_projectEstablish a container to organise jobs and runs.
list_projectsFetch the projects belonging to your organisation.
create_jobStore a reusable, named agent task template.
run_jobTrigger an instance of a stored template with input parameters.
list_runsRetrieve historical run logs.
check_resultQuery status and retrieve the results of an active or completed task.
Get started
Connect in under a minute.
Add the remote server
In ChatGPT, Claude, Cursor or any MCP client, add a remote connector pointing at https://mcp.openhelm.ai/main/mcp. No install, no terminal, no Node setup. For clients that need a local bridge, run npx mcp-remote https://mcp.openhelm.ai/main/mcp.
Connect your OpenHelm account
Click Connect and you are sent to www.app.openhelm.ai/connect to authorise via OAuth 2.1. Your encrypted OpenHelm API key is wrapped in a signed token, the server itself stores no secrets.
Add a card to start the trial
New accounts add a card to activate a 14-day free trial (you are not billed during the trial). This keeps the shared cloud compute protected from multi-account abuse.
Run tasks from your assistant
Ask your assistant in plain English. It calls the right tool, OpenHelm spins up an isolated cloud browser sandbox, and returns evidence-backed structured results, polling check_result for longer runs.
Prefer to call the engine directly from your own product? The same primitives are available over REST.
Build on the OpenHelm API→Common questions about the General-Purpose server
Do I need to install anything?
No. OpenHelm MCP servers are remote, Streamable-HTTP servers. You add a URL as a remote connector in ChatGPT, Claude, Cursor or another MCP client. Clients without native remote support can use the lightweight npx mcp-remote bridge.
Where does the work actually run?
Not on the MCP endpoint. When a tool is called, OpenHelm starts an asynchronous agent run in an isolated E2B cloud sandbox that drives a real browser and reads official sources. The MCP layer is stateless: it holds no database and stores no secrets.
What happens with long-running tasks?
The server uses a bounded-wait pattern of about 75 seconds. Fast tasks return their structured result in one turn. Slower tasks return a task_id and your assistant calls check_result a moment later to retrieve the finished output, so connections never time out mid-run.
How is it billed?
Connecting requires an OpenHelm account with a card on file, which starts a 14-day free trial. You are not charged during the trial. After that, usage runs on your OpenHelm plan.
More servers
Explore the rest of the OpenHelm MCP suite.
Investment analysts, VCs & corporate strategy teams
Deep Research
A virtual research environment that reads SEC filings, earnings prints, and live market data, grounded in official sources.
Growth marketers, content directors & SEO specialists
SEO & GEO
A virtual SEO environment: audit rankings, draft content, and measure AI engine citations of your brand.
Founders, sales & ops teams, and AI agents
Email Outreach
A virtual inbox your agent controls, send, monitor replies, and run the whole conversation to completion.
Engineering teams, indie developers & AI coding agents
Dev Changes
A virtual coding environment: clone the repo, make the change, run tests, and open a PR.
Product, marketing & competitive-intelligence teams
Site Monitoring
Watch websites, competitors and prices; get evidence-backed diffs and briefs instead of noisy pixel alerts.
Sales, support & operations teams · soon
Phone Reach-Out
A virtual phone your agent uses to call, converse, and act on the outcome, coming soon.
Sales, support & operations teams · soon
WhatsApp Reach-Out
A virtual WhatsApp inbox your agent uses to message, handle replies, and achieve the goal, coming soon.
Operations, procurement & personal assistants · soon
Online Buying
A virtual shopper that sources products, compares options, and checks out within your guardrails, coming soon.
Want it run for you instead?
The MCP servers put the tools in your assistant. If you'd rather we run the whole workflow as a managed service, book a 20-minute call and we'll scope it.