Log inSubmit server
Run402 logo

Run402

Provision and deploy Run402 backends through MCP, including Postgres, auth, storage, functions, sites, and staged releases.

InstallRepositoryWebsite
Full README Compare alternatives Embed widget Claim this listing

Install Run402

Pick your client, then add it in one click or paste the config.

STDIONode.js
~/Library/Application Support/Claude/claude_desktop_config.json
claude_desktop_config.json
{
  "mcpServers": {
    "run402": {
      "command": "npx",
      "args": [
        "-y",
        "run402-mcp"
      ]
    }
  }
}

Settings → Developer → Edit Config, merge this into mcpServers, then fully quit and reopen Claude Desktop.

Install not yet sandbox-tested— we haven't run this command through our automated sandbox yet. Not a red flag; we're working through the catalog.

Overview

Run402 MCP server gives agents an MCP interface to Run402’s backend infrastructure, covering Postgres, authentication, storage, serverless functions, static sites, and staged deployments. Run402 exposes operational tools for bootstrapping projects, applying release specifications, checking identity and account state, and diagnosing setup issues. Its `run` tool executes TypeScript SDK snippets in a sandbox, while `docs` and `expand_result` support SDK discovery and pagination. Reach for it when an MCP-native host needs to provision or operate a Run402 project without using a cloud console.

Use cases

  • Bootstrap a Run402 project and deploy its initial backend
  • Apply staged database, function, site, and route changes
  • Inspect wallet, project, identity, allowance, and lease state
  • Run custom TypeScript SDK workflows from an MCP host
  • Retrieve paginated deployment details and execution output

Key features

  • One-call project bootstrap and deployment
  • ReleaseSpec-based backend and site deployments
  • Wallet, principal, organization, and project inspection
  • Built-in diagnostics and SDK documentation lookup
  • Sandboxed TypeScript SDK execution
  • Paged retrieval of stored tool results

Tools8

What Run402 lets your agent do. Expand a tool for its parameters.

Self-reported~240 tokens of context

Security & health

What it can reach and how reliably it runs — from our own checks, not the maintainer’s claims.

ActiveLast checked 2d ago

Recent health check succeeded.

20/20 checks healthy over the last 87d

Supply-chain signal

No high-severity advisories surfaced by our automated scan.

Critical 0High 0Medium 0Low 0

Scanned 4d ago via OSV.dev · run402-mcp (npm)

Adoption & maintenance

Factual signals from GitHub, npm, and our automated checks — not a rating.

GitHub stars24Stargazers on the source repository.
npm downloads21kPackage downloads in the last 30 days.
Last commit2d agoMost recent push to the default branch.
Tools exposed8Callable tools this server registers over MCP.
Directory activity1 viewsConfig copies, upvotes, and views on AllMCPs.

How Run402 works

What Run402 MCP server does

Run402 MCP server exposes the agent-facing operations of Run402 through Model Context Protocol. It is intended for hosts such as Claude Desktop, Cursor, Cline, and Claude Code that need to provision and manage backend resources through tool calls.

The underlying platform combines a Postgres database, REST API, user authentication, content-addressed file storage, static site hosting, serverless functions, and image generation. Deployments are paid through Run402’s supported payment flows, including x402 with USDC on Base, MPP with pathUSD on Tempo or sats over Bitcoin Lightning, and card-funded allowance. The prototype tier runs on testnet.

How it works

The MCP package acts as a thin interface over Run402’s typed SDK. up handles first-time preparation and deployment in one operation, creating or linking the required wallet, tier, project, and workspace state before applying the deployment. deploy applies a ReleaseSpec to an existing project, including database changes, functions, sites, assets, subdomains, public site paths, and route replacements.

Account and environment inspection are separate operations. status reports the wallet identity used by the server, tier and lease information, allowance, projects, and active project. whoami returns the remote principal, authenticators, organization memberships, and sign-in session grade. doctor produces an overall result with blocking issues, warnings, and individual checks.

The run tool executes a TypeScript snippet against the SDK in a sandbox. Its response includes the returned value, captured logs, and SDK calls made during execution. This provides access to SDK workflows that do not have a dedicated MCP tool. Results that are too large for one response can be retrieved with expand_result, which pages stored values from run, logs, documentation answers, or deployment details.

Setup and configuration

Run402 is distributed as an MCP package and is designed for MCP-native hosts. The repository identifies Claude Desktop, Cursor, Cline, and Claude Code as supported surfaces. Run402’s broader interface also includes a CLI and typed SDK, but those are separate ways to use the platform rather than additional MCP tools.

The platform’s normal workflow starts with a project manifest and a site entry point such as index.html, then uses up to bootstrap the missing prerequisites and deploy. The CLI documentation recommends run402 up --name ... -y for an initial deployment; MCP callers can use the corresponding up operation.

Tools and capabilities

  • up: Completes missing wallet, tier, project, and workspace-link setup, then deploys.
  • deploy: Applies a release specification covering backend, site, asset, subdomain, and route changes.
  • status: Inspects the server wallet, account allowance, tier, lease, projects, and active project.
  • whoami: Inspects the remote principal, authenticators, memberships, and session grade.
  • doctor: Runs diagnostics and separates blocking findings from warnings and checks.
  • docs: Searches or retrieves the bundled SDK reference and run primer by topic or search term.
  • run: Executes TypeScript against the SDK and returns its value, logs, and recorded SDK calls.
  • expand_result: Pages stored deployment, execution, log, or documentation results.

Limitations and notes

MCP result details may be retained for later retrieval through expand_result, rather than returned in one response. The SDK retains full results for typed callers, while CLI diagnostics are written in redacted form under .run402/diagnostics/.

Run402 uses principal-based identity and explicit authority. Organization memberships, roles, grants, grant keys, freshness, and spending policy affect what an acting principal can do. The platform is open source at the agent-surface level under MIT; its self-hostable runtime slice is maintained separately under Apache-2.0, while the managed Cloud control plane is proprietary.

Read the full README View source on GitHub

Reviews

No reviews yet — be the first to share how it worked for you.

Frequently Asked Questions about Run402

Run402 is an MCP server that connects AI agents to Run402 backend infrastructure, including Postgres, authentication, storage, serverless functions, and site hosting. Its main tools bootstrap and deploy projects, inspect wallet and identity state, diagnose setup, run TypeScript SDK snippets, search bundled docs, and page large results.

Install and configure the Run402 MCP package in an MCP-compatible host. The repository identifies Claude Desktop, Cursor, Cline, and Claude Code as supported clients. Run402 also provides separate CLI and SDK packages, but those are not required when using the MCP interface.

Run402 supports Claude Desktop, Cursor, Cline, and Claude Code. The repository lists these hosts for the `run402-mcp` surface, where core Run402 operations are exposed as MCP tools.

Call `up` to complete missing wallet, tier, project, and workspace-link setup before deploying, or call `deploy` to apply a `ReleaseSpec` to an existing project. Release specifications can include database changes, functions, sites, assets, subdomains, public paths, and route replacements.

Use the `run` tool with a TypeScript snippet. It executes the snippet against the Run402 SDK in a sandbox and returns the value, captured logs, and SDK calls made. Use `expand_result` when the stored value or logs need to be retrieved in pages.

Related MCP servers

All in Databases All alternatives
Add a badge to your READMEShow it’s listed — live SVG, Markdown or HTML.

AllMCPs Directory Badge

Full Badge Customizer

Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.

Badge Style:
Live Dynamic SVG PreviewRun402 AllMCPs Directory Badge
Markdown (GitHub README)
[![AllMCPs](https://allmcps.com/api/badge/run402?style=directory)](https://allmcps.com/mcp/run402)
HTML Embed
<a href="https://allmcps.com/mcp/run402"><img src="https://allmcps.com/api/badge/run402?style=directory" alt="Run402 on AllMCPs" /></a>