@floopfloop/mcp
The @floopfloop/mcp server integrates FloopFloop's AI-powered web project platform with LLM hosts, enabling agents to create, monitor, refine, and manage web projects through natural-language instructions.
Project Lifecycle: List all projects (
list_projects), fetch by ID or subdomain (get_project), create from a prompt (create_project), refine with follow-up instructions (refine_project), check build/deploy status (project_status), block until live (wait_for_live), cancel a build (cancel_project), reactivate a cancelled project (reactivate_project), and view message timelines (get_conversations).Subdomain Management: Check slug availability (
check_subdomain) or generate a friendly slug from a prompt (suggest_subdomain).Secrets Management: List secret key names (
list_secrets), create or overwrite a secret (set_secret), or delete a secret (remove_secret). Values are one-way encrypted and never returned.File Uploads: Read a local file, upload it to S3, and get an attachment reference for use in refinements (
upload_from_path).Library Projects: Browse the public project library (
list_library_projects) and clone templates into your account (clone_library_project).Account & Billing: View user info (
whoami), current subscription and credits (current_subscription), usage summary (usage_summary), and manage API keys — list, create (secret returned once), or revoke (list_api_keys,create_api_key,remove_api_key).Safe by Design: Destructive actions (cancel, set/remove secrets, revoke keys) are marked so hosts can prompt for user confirmation.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@@floopfloop/mcpcreate a landing page for a vegan bakery called Green Bite"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
@floopfloop/mcp
Model Context Protocol server for the FloopFloop API — lets Claude Desktop, Cursor, Zed, and any other MCP-aware LLM host build, poll, and refine FloopFloop projects on the user's behalf.
Wraps the official @floopfloop/sdk and exposes a curated subset of its surface as MCP tools.
Tools
Tool | What it does |
| List every project you have access to |
| Fetch a project by id or subdomain |
| Cheap status snapshot — safe to poll |
| Kick off a new build from a natural-language prompt |
| Send a refinement message to an existing project |
| Block until the project reaches a terminal state |
| Stop a queued or in-progress build |
| Resume a cancelled / archived project |
| Read a project's message timeline (prompts + replies + deploy markers) |
| Is a given slug available? |
| Generate a friendly slug from a prompt |
| List secret keys on a project (names only) |
| Create/overwrite a project secret |
| Delete a project secret |
| Browse the public project library |
| Duplicate a library project into the user's account |
| Plan limits + current-period credit / build / storage usage |
| List the user's API keys (never returns the raw secret) |
| Mint a new API key — raw secret returned ONCE |
| Revoke an API key by id or name |
| Read a local file, presign + upload to S3, return an attachment ref |
| Show the authenticated user |
| Plan tier + credit balance (price, billing period, current/rolled-over credits) |
Configuration
Grab an API key: floop keys create mcp-host (via the floop CLI) or the dashboard → Account → API Keys. Business plan required to mint new keys.
The server reads FLOOP_API_KEY from the environment. Optionally override FLOOP_API_URL to point at staging.
Claude Desktop
~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %AppData%\Claude\claude_desktop_config.json (Windows):
{
"mcpServers": {
"floopfloop": {
"command": "npx",
"args": ["-y", "@floopfloop/mcp"],
"env": { "FLOOP_API_KEY": "flp_..." }
}
}
}Restart Claude Desktop; the tools show up under the 🔌 icon.
Cursor
~/.cursor/mcp.json (or Cursor Settings → Tools & Integrations → MCP):
{
"mcpServers": {
"floopfloop": {
"command": "npx",
"args": ["-y", "@floopfloop/mcp"],
"env": { "FLOOP_API_KEY": "flp_..." }
}
}
}Generic MCP host
Any host that speaks the MCP stdio transport can run the server:
FLOOP_API_KEY=flp_... npx -y @floopfloop/mcpInstall
The package is pulled in automatically via npx -y @floopfloop/mcp in the config snippets above. If you prefer a pinned install:
npm install -g @floopfloop/mcp
which floop-mcpUsage notes
create_projectstarts a build right away. Follow up withwait_for_live(blocks) orproject_status(poll) to know when it's up.wait_for_livedefaults to a 10-minute ceiling (bounded bytimeoutMs, capped at 30 min). Most builds finish in under two minutes.upload_from_path→refine_projectis the canonical attachment flow: callupload_from_pathwith a local file, get back anUploadedAttachment({key, fileName, fileType, fileSize}), then pass it asattachments: [<that object>]onrefine_project. The LLM host's process needs read access to the file path; max 5 MB.refine_projectcodeEditOnly: trueruns a 3-step in-place patch instead of a full 6-step rebuild and charges roughly half the credits — use it for copy edits, colour swaps, or typo fixes on a project that's already live. The backend won't promote a code-edit to a full refinement automatically, so prefer plainrefine_projectwhen the change actually needs redesign.cancel_project→reactivate_projectis the abort/redo pattern.cancel_projectisdestructiveHint: true; hosts should confirm before calling it.reactivate_projecttriggers a fresh build at the project's most recent prompt.set_secret/remove_secretare markeddestructiveHint: true— hosts may ask the user to confirm before they run.list_secretsonly returns names, never values. Secret values cannot be retrieved once written; rotate them by re-setting.On failure, tools return an MCP
isErrorcontent result rather than tearing down the session, so the host displays the error to the user.
Development
git clone https://github.com/FloopFloopAI/floop-mcp.git
cd floop-mcp
npm install
npm run typecheck
npm run build
FLOOP_API_KEY=flp_... node dist/index.jsTo smoke-test the stdio handshake without an LLM host:
printf '%s\n' \
'{"jsonrpc":"2.0","id":1,"method":"initialize","params":{"protocolVersion":"2024-11-05","capabilities":{},"clientInfo":{"name":"smoke","version":"0"}}}' \
'{"jsonrpc":"2.0","method":"notifications/initialized"}' \
'{"jsonrpc":"2.0","id":2,"method":"tools/list"}' \
| FLOOP_API_KEY=flp_dummy node dist/index.js 2>/dev/nullReleasing
.github/workflows/release.yml publishes to npm with provenance whenever
a tag matching mcp-v* is pushed. One-time setup before the first
release:
Create an automation npm token on the
@floopfloopscope (publishes bypass 2FA by design when tokens are marked automation).Add it as the repo secret
NPM_TOKEN:gh secret set NPM_TOKEN --repo FloopFloopAI/floop-mcp # paste token when prompted — do NOT use --body, which can leak it to # shell history.Tag + push:
npm version 0.1.0-alpha.1 --no-git-tag-version # if needed git tag mcp-v$(node -p "require('./package.json').version") git push --follow-tags
The workflow typechecks, builds, runs the stdio smoke test, verifies the
tag matches package.json, then publishes and cuts a GitHub Release
(prerelease, auto-generated notes).
License
MIT
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Latest Blog Posts
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/FloopFloopAI/floop-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server