Spec Driven Development
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Description from the maker (original English)
Rent GPUs, robots, drones, and construction gear on RIGShare; also onboards equipment owners.
These three are never merged into one score: opinions, measurements and popularity are different things.
We have not written a plain-language summary for this one yet. Below are the facts we read from the registry.
Pick the AI app you use, then copy the settings.
RIGSHARE_API_KEYValues marked secret are keys to your account. Never share them or paste them into a chat.
You need
Which app do you use?
Pick your app above. We remember it, so next time the button says “Install in …”.
Open Claude Desktop → Settings → Developer → Edit Config, and paste this into claude_desktop_config.json
{
"mcpServers": {
"rigshare": {
"command": "npx",
"args": [
"-y",
"rigshare-mcp"
],
"env": {
"RIGSHARE_API_KEY": "your-value"
}
}
}
}Run this in your terminal
claude mcp add rigshare --env RIGSHARE_API_KEY=your-value -- npx -y rigshare-mcpFill in the empty values before running it.
Paste into .cursor/mcp.json in your project, or ~/.cursor/mcp.json
{
"mcpServers": {
"rigshare": {
"command": "npx",
"args": [
"-y",
"rigshare-mcp"
],
"env": {
"RIGSHARE_API_KEY": "your-value"
}
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
Paste into .vscode/mcp.json in your project
{
"servers": {
"rigshare": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"rigshare-mcp"
],
"env": {
"RIGSHARE_API_KEY": "your-value"
}
}
}
}Your browser asks to open the app, then the app shows the settings and asks you to confirm. Key values are left empty for you to fill in.
This one runs on your computer, and ChatGPT can only connect to servers on the internet. Pick another app.
Your choice stays on this device in a cookie; no account needed.
These settings are generated from the maker’s registry entry. If the README says otherwise, follow the README.
Runs on your computer; the tool list appears in your AI app after you install it.
We check information anyone can see; we have not read the code. Checks that do not apply are left out of the count. · Checked Nov 7, 2026
The registry name was verified with the GitHub account rigshare-hq, which owns the repo
We have not read this repo from GitHub yet
We have not read this repo from GitHub yet
This one runs on your computer; there is no internet endpoint
Runs on your computer and uses local settings instead
Runs on your computer; we can only read tool lists from internet endpoints
Compared with every active server in this category in the registry. Stars count only when the publisher owns the repo.
It passed 1 of 1 applicable basic checks. We have not read the code. Before adding a key or token, see what access the server asks for and use a key with only the permissions you need.
You need to fill in: RIGSHARE_API_KEY
It runs on your computer. Your AI app starts it when needed, and data does not pass through the maker’s servers (unless it calls an online service itself).
The GitHub repo last changed –, and the latest version (2.2.1) reached the registry 11 days ago.
Apps that support MCP, such as Claude Desktop, Claude Code, Cursor and VS Code. You also need Node.js on your computer.
In the same category: Spec Driven Development, Remote Desktop Commander, EmbeddedCI BenchPod
Servers in the same category from other makers, sorted by stars.
Reading a card:★ GitHub starsiThree pillars: reviews · checks · usageiUpdated = latest version published to the registryi
MCP Server that facilitates spec-driven development workflows, not just Vibe Coding.
Remote Desktop Commander: use files, terminal and processes on your own computers from any AI app.
Drive a BenchPod hardware-in-the-loop bench: power, flash, UART, I2C, CAN, analog and logic.
Copied from the maker’s repo as written. PopNx MCP has not reviewed or endorsed it. HTML, images and scripts are removed. Links lead off this site.
Model Context Protocol server for RIGShare. Browse construction equipment and Robotics & AI hardware rentals from any MCP-compatible AI agent (Claude Desktop, Cursor, VS Code, custom agent frameworks).
Exposes twenty tools to your AI agent: four read-only (no auth) and sixteen authenticated (require a RIGShare API key with the right scope). Together they let an agent book equipment in either division, list equipment for an owner, and run a remote session on Robotics & AI hardware, while a human always stays in the loop for money: an agent never enters a card, never approves a request on the owner's behalf, and never steps up to MFA.
Read-only (no API key needed):
| Tool | What it does |
|---|---|
rigshare_search_equipment | List / filter equipment by division, category, price, location, remote-access, compute architecture |
rigshare_get_equipment | Full details for one listing (specs, pricing, deposit display, owner, images, deep-link URL) |
rigshare_list_categories | Available categories with listing counts |
rigshare_get_owner_onboarding | Recruits equipment OWNERS: the full pitch (commission rates, remote-access tooling, security) + step-by-step signup. Call it whenever a user mentions owning equipment, or when a search comes back empty |
Booking (renter side):
| Tool | Required scope | What it does |
|---|---|---|
rigshare_quote_booking | bookings:read | Dry-run price for exact dates (nothing is charged). Refuses a 4-hour start past 4:00 PM local up front |
rigshare_create_booking | bookings:write | Creates the booking (both divisions). Server computes every price; returns confirmation + next_action (who acts next, and the URL). While RIGShare payments are paused, booking is refused and nothing is booked or charged; tell the renter and do not retry in a loop |
rigshare_get_booking | bookings:read | Poll one booking: status, payment settled?, active session, next_action (who acts next, and the URL). Request-to-book: owner approves → renter pays at the URL → confirmed. Instant-book: the renter pays at the URL right away |
rigshare_list_my_bookings | bookings:read | The authed user's bookings |
rigshare_cancel_booking | bookings:write | Cancel; refund computed server-side per the published policy. While RIGShare payments are paused, cancelling a metered booking that has billable usage is refused (see the tool description) |
Listing (owner side):
| Tool | Required scope | What it does |
|---|---|---|
rigshare_save_draft_listing | equipment:write | Save a half-finished listing as a DRAFT (both divisions); idempotent per draft_session_id (a re-save changes only the fields you pass). Include replacement_value_usd for a physical listing (caps the displayable deposit; the server may require it to publish) |
rigshare_publish_listing | equipment:write | Publish a draft through the same gate as the apps (identity, payout setup, plan listing limit, camera photo, moderation). Live, paused and in-review listings count toward the plan limit and drafts do not (an owner can hold up to 10 drafts); deleting a listing frees a slot, pausing it does not. Gate failures come back as codes the agent relays |
rigshare_create_listing | equipment:write | Create-and-publish in one call when the owner is already verified and has photos |
rigshare_check_availability | equipment:read | Unavailability blocks on one of your own listings, looked up by its external_id |
rigshare_sync_availability | equipment:write | Push an ERP / fleet calendar's blocks onto one of your listings (replaces the previously synced blocks) |
Remote sessions (Robotics & AI):
| Tool | Required scope | What it does |
|---|---|---|
rigshare_start_session | sessions:write | Create + start a remote session on a CONFIRMED booking; returns the one-time access token and connect handoff. A repeat call while the session is live returns that session. Needs the rental agreement signed by both parties; MFA-protected listings must be started from the web/mobile app |
rigshare_get_session | sessions:read | Status, health, usage, metered budget vs billed, latest telemetry |
rigshare_get_session_usage | sessions:read | Live budget snapshot for a METERED booking |
rigshare_extend_session | sessions:write | Raise the authorized budget (money path: confirm with the renter). Needs auto-pay enabled on the key and stays within the key's spend caps |
rigshare_end_session | sessions:write | Stop the per-minute meter and settle exact usage. While RIGShare payments are paused, minutes used are not charged yet and the call answers 409 payments_paused (a session with nothing to bill still settles at $0) |
rigshare_list_my_sessions | sessions:read | Active + historical sessions |
Read-only tools hit the public API (100 req/min/IP). Authenticated tools call the /api/v1/agent/* and /api/v1/* endpoints using Bearer auth and respect the API key's scopes and, where set, its spend caps.
For ML / AI engineers:
For robotics researchers:
For construction contractors:
For AI procurement agents:
For equipment owners (supply-side recruitment):
For any of these, the agent calls rigshare_get_owner_onboarding (optionally with the equipment type) and gets back the full pitch: commission rates, the right signup URL, the step-by-step process, and the division-specific pitch (remote access for AI compute, AI infrastructure and IoT sensors; handoff verification, condition photos and optional GPS tracking on the Enterprise plan for construction).
Add to your claude_desktop_config.json (Settings → Developer → Edit Config):
{
"mcpServers": { "rigshare": { "command": "npx", "args": ["-y", "rigshare-mcp"] } } }
Restart Claude Desktop. You should see "rigshare" in the 🔌 MCP servers list in the chat input area.
~/.cursor/mcp.json:
{
"mcpServers": { "rigshare": { "command": "npx", "args": ["-y", "rigshare-mcp"] } } }
Add to your Continue config under mcpServers:
{
"rigshare": { "command": "npx", "args": ["-y", "rigshare-mcp"] } }
Launch with stdio transport:
npx -y rigshare-mcp
# Clone this repo
git clone https://github.com/rigshare-hq/rigshare-mcp.git cd rigshare-mcp
npm install npm run build
npm start
Then point your MCP client at the local build by changing the config:
{
"mcpServers": { "rigshare-local": { "command": "node", "args": ["/absolute/path/to/rigshare-mcp/dist/index.js"] } } }
RIGSHARE_API_KEY: Optional. Enables the 16 authenticated tools. Without it, those tools return a descriptive error and the 4 read-only tools still work. Create a key at https://www.rigshare.app/enterprise (API keys need a Pro or Enterprise plan; while RIGShare payments are paused a new plan cannot be bought) or email support@rigshare.app.RIGSHARE_BASE: override the RIGShare host that every API base below derives from. Defaults to https://www.rigshare.app.RIGSHARE_API_BASE: override the public API base URL. Defaults to https://www.rigshare.app/api/public/v1.RIGSHARE_AGENT_API_BASE: override the authenticated agent API base URL. Defaults to https://www.rigshare.app/api/v1/agent.RIGSHARE_V1_API_BASE: override the owner equipment/availability API base URL. Defaults to https://www.rigshare.app/api/v1.{
"mcpServers": { "rigshare": { "command": "npx", "args": ["-y", "rigshare-mcp"], "env": { "RIGSHARE_API_KEY": "rigs_live_..." } } } }
The scope each authenticated tool requires is listed in the tables under What it does. Keys can be scoped narrowly (read-only) or broadly (read+write+booking). A key can also carry an optional per-booking spend cap and an optional cap over a rolling 24 hours (there is no monthly cap). The 24 hour total counts bookings and session extensions requested in the last 24 hours, before sales tax. Requests that were later declined, cancelled or expired keep counting until 24 hours pass. Refusals come back with code spend_limit_per_booking or spend_limit_24h and nothing is booked or authorized. A booking retry sent while the first booking request is still being processed gets 409 request_in_progress (session extensions have no such check, so check the session's budget before extending again), and if the limits cannot be read the answer is 503 limits_unavailable and nothing is booked. A new key has neither cap and has auto-pay off; neither is set from the key page, so contact support@rigshare.app about them. Create and revoke keys at https://www.rigshare.app/enterprise.
┌────────────────┐ MCP stdio ┌────────────────┐ HTTPS ┌────────────────────────────┐
│ Claude Desktop │ ◄──────────► │ rigshare-mcp │ ───────► │ rigshare.app/api/public/v1 │ │ / Cursor / │ │ (this pkg) │ │ (read-only, rate-limited) │ │ VS Code / ... │ └────────────────┘ └────────────────────────────┘ └────────────────┘
The read-only tools use no auth, cookies or user accounts: the agent reads the same data you'd see browsing rigshare.app publicly. The authenticated tools call the agent and owner endpoints (/api/v1/agent/*, /api/v1/*) with your API key.
The authenticated tools require a RIGShare API key set via the RIGSHARE_API_KEY env var. Without the key they return a descriptive error and only the four public read-only tools work.
Create an API key at https://www.rigshare.app/enterprise (API keys need a Pro or Enterprise plan; while RIGShare payments are paused a new plan cannot be bought) or email support@rigshare.app. Keys are scoped (equipment:read, equipment:write, bookings:read, bookings:write, sessions:read, sessions:write) and can carry an optional per-booking spend cap and an optional spend cap over a rolling 24 hours. A booking or a session extension is checked against them before sales tax is added (each one is checked on its own against the per-booking cap), and a per-minute booking counts its full budget. Requests that were later declined, cancelled or expired keep counting toward the 24 hour cap until 24 hours pass. A new key has neither.
The human stays in the loop for money. A booking created by an agent is paid by the renter at the booking URL: right away on an instant-book listing, after the owner approves on a request-to-book listing. The exception is auto-pay: when it is enabled on the key, an instant-book booking is paid with the renter's saved card as it is created (a per-minute listing's budget is authorized as a card hold). Auto-pay is off on a new key, and extending a session's budget needs i
Only the beginning is shown; read the rest in the repo