Regen Compute
Enables AI coding assistants to estimate session energy footprint and retire verified ecocredits on Regen Network, providing on-chain proof of regenerative contribution.
README
Regen Compute
Add Regenerative AI to your AI coding assistant in 30 seconds.
Every AI session consumes energy. This MCP server lets your AI assistant estimate that footprint and retire verified ecocredits on Regen Network — with immutable on-chain proof.
This is regenerative contribution, not carbon offsetting. No neutrality claims. Just verified funding of ecological regeneration.
Install
claude mcp add -s user regen-compute -- npx regen-compute
That's it. Works immediately — no API keys, no wallet, no configuration needed for read-only tools.
Supported Platforms
<details> <summary><b>Cursor</b></summary>
Add to ~/.cursor/mcp.json (global) or .cursor/mcp.json (project):
{
"mcpServers": {
"regen-compute": {
"command": "npx",
"args": ["regen-compute"]
}
}
}
Restart Cursor. Tools appear in Cursor's AI chat.
</details>
<details> <summary><b>Windsurf</b></summary>
Add to ~/.codeium/windsurf/mcp_config.json:
{
"mcpServers": {
"regen-compute": {
"command": "npx",
"args": ["regen-compute"]
}
}
}
Open Windsurf Settings > Cascade > MCP Servers to verify the connection.
</details>
<details> <summary><b>VS Code (GitHub Copilot)</b></summary>
Add to .vscode/mcp.json in your workspace:
{
"servers": {
"regen-compute": {
"command": "npx",
"args": ["regen-compute"]
}
}
}
Requires VS Code 1.99+ with GitHub Copilot. Tools are available in Agent mode.
</details>
<details> <summary><b>JetBrains (IntelliJ, WebStorm, PyCharm, etc.)</b></summary>
Go to Settings > Tools > AI Assistant > Model Context Protocol (MCP), click +, and paste:
{
"mcpServers": {
"regen-compute": {
"command": "npx",
"args": ["regen-compute"]
}
}
}
Requires a JetBrains AI Assistant subscription.
</details>
<details> <summary><b>Gemini CLI</b></summary>
Add to ~/.gemini/settings.json:
{
"mcpServers": {
"regen-compute": {
"command": "npx",
"args": ["regen-compute"]
}
}
}
</details>
<details> <summary><b>Continue.dev</b></summary>
Create .continue/mcpServers/regen-compute.yaml in your workspace:
name: Regen Compute
version: 0.0.1
schema: v1
mcpServers:
- name: regen-compute
type: stdio
command: npx
args:
- "regen-compute"
Note: MCP tools are only available in Continue's agent mode.
</details>
<details> <summary><b>Sourcegraph Cody</b></summary>
Cody uses the OpenCtx bridge, which requires a local file path (not npx). Install first:
npm install -g regen-compute
Then add to your VS Code settings.json:
{
"openctx.providers": {
"https://openctx.org/npm/@openctx/provider-modelcontextprotocol": {
"nodeCommand": "node",
"mcp.provider.uri": "file:///ABSOLUTE/PATH/TO/node_modules/regen-compute/dist/index.js"
}
}
}
Replace the path with the actual location of the installed package.
</details>
<details> <summary><b>Any MCP-compatible client</b></summary>
Most MCP clients accept this standard config:
{
"mcpServers": {
"regen-compute": {
"command": "npx",
"args": ["regen-compute"]
}
}
}
</details>
ChatGPT and OpenAI API: These platforms require a remote HTTP MCP server (streamable HTTP or SSE transport). Regen Compute currently uses stdio transport. HTTP transport support is on the roadmap.
Quick Start
After installing, ask your AI assistant to run these tools:
estimate_session_footprint— See the energy and CO2 cost of your current sessionbrowse_available_credits— Explore live ecological credits (carbon, biodiversity, marine) on Regen Marketplaceretire_credits— Retire credits to fund regeneration (credit card link by default, or on-chain with a wallet)get_retirement_certificate— Get a verifiable on-chain proof of your contribution
Or just ask: "Estimate my session's footprint and show me credits I can retire" — the AI handles the rest.
What You Get
| Tool | What it does |
|---|---|
estimate_session_footprint |
Estimate energy, CO2, and retirement cost for your AI session |
browse_available_credits |
Browse live sell orders on Regen Marketplace (carbon, biodiversity, marine, species) |
retire_credits |
Retire credits on-chain (with wallet) or get a credit card purchase link (without) |
get_retirement_certificate |
Verify any retirement with on-chain proof |
get_impact_summary |
Regen Network aggregate stats — projects, retirements, jurisdictions |
browse_ecobridge_tokens |
List 50+ tokens across 10+ chains for cross-chain payment |
retire_via_ecobridge |
Pay with USDC/ETH/etc. on Ethereum, Base, Polygon, Arbitrum, and more |
Three payment modes — pick what works for you
- Credit card (default, no setup) — get a Regen Marketplace link
- Direct on-chain — set
REGEN_WALLET_MNEMONIC, retire in a single tx - Any token, any chain — USDC on Base, ETH on Arbitrum, etc. via ecoBridge
How It Works
Your AI Assistant (Claude Code / Cursor / etc.)
│
│ MCP Protocol (stdio)
▼
Regen Compute MCP Server
│
├─ Footprint estimation (heuristic, clearly labeled as approximate)
├─ Credit browsing (live sell order data from Regen Ledger)
├─ Retirement execution (MsgBuyDirect with auto-retire)
└─ Certificate retrieval (on-chain verification)
│
▼
Regen Network Ledger (immutable, verifiable, non-reversible)
If anything fails during on-chain retirement, it falls back to a marketplace link. Users are never stuck.
Configuration
Works with zero config. Set environment variables for advanced features:
cp .env.example .env
| Variable | Required | What it enables |
|---|---|---|
| (none) | — | Footprint estimation, credit browsing, impact stats, marketplace links |
REGEN_WALLET_MNEMONIC |
Optional | Direct on-chain retirement (MsgBuyDirect) |
ECOBRIDGE_EVM_MNEMONIC |
Optional | Cross-chain payment via ecoBridge (send USDC, ETH, etc.) |
ECOBRIDGE_ENABLED=false |
Optional | Disable ecoBridge tools |
See .env.example for all options with inline documentation.
Developer Resources
MCP Tools
| Tool | Description | Auth required |
|---|---|---|
estimate_session_footprint |
Estimate AI session energy use and CO2 | No |
estimate_monthly_footprint |
Personalized monthly footprint with location/product multipliers | No |
browse_available_credits |
Live sell orders from Regen Marketplace | No |
retire_credits |
Retire ecocredits on-chain or via marketplace link | Wallet |
get_retirement_certificate |
Retrieve on-chain retirement certificate by nodeId or txHash | No |
get_impact_summary |
Regen Network aggregate impact statistics | No |
check_subscription_status |
Subscriber status, cumulative impact, referral link | API key |
get_regen_price |
REGEN/USD price from CoinGecko with cache status | No |
verify_payment |
Verify on-chain payment across 19 chains (16 EVM + BTC + SOL + TRX) | No |
get_community_goals |
Community retirement goals, progress, subscriber count | No |
browse_ecobridge_tokens* |
Cross-chain tokens and prices via ecoBridge | No |
retire_via_ecobridge* |
Retire credits by sending tokens on any EVM chain | EVM wallet |
* Requires ECOBRIDGE_ENABLED=true
MCP Prompts
Pre-built workflows you can invoke:
| Prompt | Workflow |
|---|---|
offset_my_session |
Estimate footprint → browse credits → retire |
show_regen_impact |
Pull live network stats and summarize |
retire_with_any_token* |
Browse ecoBridge tokens → pick chain/token → retire |
* Requires ECOBRIDGE_ENABLED=true
CLI Subcommands
| Command | Description |
|---|---|
npx regen-compute |
Start MCP server (stdio transport) |
npx regen-compute serve |
Start the web server (port 3141) |
npx regen-compute pool-run [--dry-run] |
Execute monthly pool retirement batch |
REST API Endpoints
All endpoints prefixed with /api/v1/. Protected endpoints require Authorization: Bearer <api_key>.
| Endpoint | Method | Auth | Description |
|---|---|---|---|
/openapi.json |
GET | No | OpenAPI 3.1 specification |
/payment-info |
GET | No | Payment addresses and pricing tiers |
/confirm-payment |
POST | No | Confirm crypto payment for subscription provisioning |
/retire |
POST | Yes | Retire ecocredits |
/credits |
GET | Yes | Browse available credits |
/footprint |
GET | Yes | Estimate session footprint |
/certificates/:id |
GET | Yes | Get retirement certificate |
/impact |
GET | Yes | Network impact statistics |
/subscription |
GET | Yes | Subscription status and cumulative impact |
/community/goals |
GET | Yes | Community goals, progress, and subscriber stats |
/scheduled-retirements |
GET | Yes | Scheduled retirements with status breakdown |
Credit Types
| Type | Description |
|---|---|
| Carbon (C) | Verified carbon removal and avoidance |
| Biodiversity (BT) | Terrasos voluntary biodiversity credits |
| Marine Biodiversity (MBS) | Marine ecosystem stewardship |
| Umbrella Species (USS) | Habitat conservation via umbrella species |
| Kilo-Sheep-Hour (KSH) | Grazing-based land stewardship |
Data Sources
| Source | What it provides |
|---|---|
| Regen Ledger REST | Credit classes, projects, batches, sell orders |
| Regen Indexer GraphQL | Retirement certificates, marketplace orders, stats |
| Regen Marketplace | Credit card purchase flow |
| ecoBridge API | Cross-chain tokens, prices, widget links |
Development
Local setup
git clone https://github.com/regen-network/regen-compute.git
cd regen-compute
npm install
cp .env.example .env # fill in your keys (optional — read-only tools work without)
npm run build
Connect to Claude Code (local build)
Point your MCP config at the local build so changes are reflected immediately:
claude mcp add regen-compute -s user -- node /path/to/regen-compute/dist/index.js
To enable ecoBridge tools locally:
claude mcp add regen-compute -s user \
--env ECOBRIDGE_ENABLED=true \
-- node /path/to/regen-compute/dist/index.js
Scripts
| Command | What it does |
|---|---|
npm run dev |
Watch mode with hot reload (tsx) |
npm run build |
Production build (tsc) |
npm run typecheck |
Type checking only |
npm test |
Run test suite (vitest) |
npm run test:watch |
Run tests in watch mode |
Running tests
npm test # single run
npm run test:watch # re-run on file changes
Tests mock all external APIs (RPC providers, ecoBridge, Regen Ledger) — no network calls, no wallets needed.
Roadmap
| Phase | Status |
|---|---|
| 1.0 Read-only MCP — footprint, browsing, marketplace links, certificates | Complete |
| 1.5 On-chain retirement — wallet signing, order routing, ecoBridge cross-chain | Complete |
| 2.0 Subscription pool — Stripe, monthly batch retirements, fractional attribution | In progress |
| 3.0 Smart contract — CosmWasm pool, automated retirement, REGEN burn | Planned |
| 4.0 Scale — enterprise API, platform partnerships, credit supply development | Planned |
See ROADMAP.md for the full 3-track rollout plan, dependency graph, and business context.
Contributing
We welcome contributions! See CONTRIBUTING.md for the issue workflow, label guide, and branch conventions.
Good first issues: gh issue list --label "good first issue"
License
Apache-2.0 — see LICENSE.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.