VibeKit.bot

Already added avibekitserver, or rotating your key?Prefer the header form. Because the?api_key=URL bakes the key into the endpoint, changing keys means replacing the whole server entry, and some clients keep the old one until you explicitly overwrite it, so you can end up authenticating with a stale key without noticing. With the bare URL +Authorization: Bearerheader (or theVIBEKIT_API_KEYenv var for local/stdio), the endpoint never changes, so rotating a key is just swapping the header value. If your client added the URL form, delete the existingvibekitentry before re-adding rather than adding a second one.

Prefer running it locally (e.g. Claude Desktop, Cursor)?

Edit~/Library/Application Support/Claude/claude_desktop_config.json(macOS) or the equivalent on Windows / Linux:

{ "mcpServers": { "vibekit": { "command": "vibekit-mcp", "env": { "VIBEKIT_API_KEY": "vk_live_..." } } } }

Restart Claude Desktop. The new tools appear under the connectors icon.

Cursor → Settings → Features → Model Context Protocol → Add a new MCP server. Same command + env vars.

36 tools exposed today, grouped by surface:

List all hosted apps in your VibeKit account, each with its id, subdomain, status, and live URL. Start here to get an appId for the other app tools.

Get full details for one hosted app — status, subdomain/URL, repo, runtime, and database/domain config. Use the appId from vibekit_list_apps.

Create a new hosted app from a starter template and deploy it live at <subdomain>.vibekit.bot. Call vibekit_list_templates first to pick a valid template. To deploy existing code from a GitHub repo instead, use vibekit_deploy.

List the starter templates available to vibekit_create_app (e.g. landing, dashboard, blog, saas, crud-api). Call this before vibekit_create_app to choose a valid template.

Deploy a GitHub repo to VibeKit hosting, live at <subdomain>.vibekit.bot (Fargate container behind nginx, SSL automatic). Use when the code already lives in a GitHub repo; for a starter template instead use vibekit_create_app.

Redeploy an existing app to pull and ship the latest code. Use after pushing changes to the app's GitHub repo.

Fetch recent container/application logs for an app. Use this to debug crashes, errors, or a deploy that isn't serving.

Restart a hosted app's container. Use when an app is wedged, or to apply changed environment variables.

Stop a hosted app's container. The app stays in your account and can be brought back with vibekit_start_app. Use to take an app offline without deleting it.

Start a stopped hosted app's container. Use to bring an app stopped via vibekit_stop_app back online.

Get the environment variables configured for an app (names and values). Read-only — use vibekit_set_env to change them.

Set or update environment variables for an app (merges with existing). The app must be restarted or redeployed for changes to take effect.

Permanently delete a hosted app — stops the container and removes its workspace, database, env vars, and subdomain. Irreversible (the GitHub repo is NOT touched). Confirm with the user before calling.

List an app's recent deploys, newest first, with status and commit info. Find a deployId here to pass to vibekit_rollback_deploy.

Roll an app back to a previous deploy (get the deployId from vibekit_list_deploys). The container restarts on the rolled-back build; reversible by rolling forward again.

Send a message to an app's built-in AI coding agent, which reads, writes, and modifies the app's code and redeploys it. Use for 'build/make a change to my app' requests. If the agent finishes quickly you get its reply directly. Otherwise you get status:'working' — do NOT resend; instead give the user LIVE progress: poll vibekit_agent_status every few seconds and relay the current step from activity.status ('editing the homepage…', 'deploying…') until activity.done, then read vibekit_agent_history for the final reply.

Check whether an app's AI agent is idle or working, and get its LIVE progress. Returns activity.status = the current human-readable build step (e.g. 'editing index.html', 'deploying'), activity.steps = the steps so far, and activity.done = whether the turn finished. Poll this every few seconds while a build is running and relay activity.status to the user so they see progress. Once activity.done is true the response also includes liveCheck = a server-side fetch of the deployed page ({ url, ok, httpStatus, title }); use it to confirm the app is live and renders — do NOT curl/fetch the URL yourself, your sandbox may block it.

Get the recent chat history between the user and an app's AI agent.

Provision a Postgres database for an app (idempotent — safe if already enabled). Required before the vibekit_db_* tools work. Note: free-tier apps need the $3/mo Database add-on.

Get an app's database status and connection info (whether it's provisioned, frozen, etc.).

Get an app's database schema — every table and its columns. Always call this before writing a query with vibekit_db_query.

Run a read-only SQL query against an app's Postgres database and return up to 200 result rows. SELECT only — writes and DDL (INSERT/UPDATE/DELETE/ALTER/DROP/…) are rejected server-side; use vibekit_chat or vibekit_submit_task to have the agent make data or schema changes. Call vibekit_db_schema first to learn the tables. SQL string, max 5000 chars.

Browse rows of one table with pagination and optional sorting — a quick read-only alternative to hand-writing a SELECT.

Kick off an automated QA run (headless-browser checks) against a hosted app. Poll vibekit_qa_status for results.

Get the latest automated QA results and status for an app.

Submit an autonomous coding task. The AI writes the code, commits to GitHub, and (by default) deploys to <subdomain>.vibekit.bot. Returns a taskId — poll vibekit_get_task or block with vibekit_wait_for_task. Use for 'build X' or 'change X' work against a repo.

Get the status and result of a task submitted via vibekit_submit_task.

List recent coding tasks, optionally filtered by status.

Block until a submitted task finishes (or the timeout), polling every 5 seconds, then return its result. Use right after vibekit_submit_task when you want the final result inline.

Cancel a running task by taskId. No-op if it has already finished.

Create a recurring coding task that the AI runs automatically on a cron schedule (e.g. a weekly SEO pass). Use for 'every Monday, do X to my repo' requests.

List all of the account's scheduled recurring tasks.

Delete a scheduled recurring task by scheduleId. Confirm with the user first.

Get VibeKit account info: plan, credit balance, and usage.

List available implementation skills (IDs, names, descriptions, tags) from the VibeKit skills registry. Discover skills here before fetching one with vibekit_get_skill.

Fetch the full content of a specific skill — implementation patterns, code examples, and best practices for a domain. Fetch on-demand when you need guidance on a specific topic.

- Stay in your editor.No tab-switching. Deploy from inside the same chat you're writing the code in.
- Compositional.Claude can chain calls, "deploy this, wait for it to come up, then set these 3 env vars, then restart", without you nursing each step.
- Scriptable from agents.If you have a Claude Code workflow that needs to deploy a sibling app, MCP makes it one tool call.

It exposes your VibeKit account as tools that any MCP client (Claude Desktop, Cursor, Zed, Continue, etc.) can call. You tell Claude Desktop 'deploy this repo and set OPENAI_API_KEY' and Claude actually does it via the MCP tools, no copy-pasting URLs into a separate dashboard.

Run npm install -g vibekit-mcp then add it to your Claude Desktop / Cursor MCP config. For Claude Desktop, edit ~/Library/Application Support/Claude/claude_desktop_config.json and add vibekit-mcp under mcpServers. The server uses your VibeKit API key for auth, get one from app.vibekit.bot/settings.

Any client that implements the Model Context Protocol spec, currently Claude Desktop, Cursor, Zed, Continue, and a growing list. The server uses stdio transport so it works with the standard MCP config block any of them accept.

Yes, the npm package is free and open-source. VibeKit's free tier (1 hosted app) is enough to start. The MCP server is just a thin RPC layer; your usage is bounded by your VibeKit plan and your client's own model costs.

This is a web browser that enables your coding agent, such as Claude Code, to visit websites on your behalf and assist you in identifying bugs or creating UI test cases.

Create crafted UI components inspired by the best 21st.dev design engineers.

Bring agent evaluations, observability, and synthetic test set generation directly into your IDE for free with Galileo's new MCP server

An MCP server to help AI assistants to answer questions and generate AccelByte Extend SDK code more effectively .

MCP server for AI Diagram Maker — generate beautiful software engineering diagrams directly inside Cursor, Claude Desktop, Claude Code, or any MCP-compatible AI agent

ALAPI MCP Tools,Call hundreds of API interfaces via MCP

AI-powered SVG animation generator that transforms static files into animated SVG components using the Allyson platform

MCP server that gives AI assistants on-demand access to 1,500+ amCharts docs, ~300 code examples, and 1000+ class API references.

APIMatic MCP Server is used to validate OpenAPI specifications using APIMatic. The server processes OpenAPI files and returns validation summaries by leveraging APIMatic’s API.

One shared context layer for AI agents and humans — live API specs, DB schemas, and versioned contracts across repos so every agent and teammate works from the same source of truth.