Model Context Protocol

The tokenomics MCP.

Plug the largest open structured tokenomics dataset on the web — 300+ real-world projects — into Claude, Cursor, Antigravity, or any MCP-aware AI assistant. Plus a 25-tool computation library (health score, peer comparison, sell-pressure projection, scenario simulation, design proposal from a brief, similarity search, risk audit, validation) using the same math the visual editor uses.

Free and open. Your first call needs no signup and no key — anonymous access works immediately (100 calls/day per IP). Want more? Grab a free key for 1,000/day. Hosted at tokeneconomics.io/api/mcp.

Install in 30 seconds↓Browse the 25 tools Endpoint URL ↗

Why use it

Three concrete scenarios. AI assistants are starting to do real work for token founders; they need real data + working math, not vibes.

INSIDE CURSOR / ANTIGRAVITY

While coding your token contract, ask your IDE "what does GMX's vesting look like?" and get structured data inline — no tab-switching to dig through whitepapers.

FROM CLAUDE DESKTOP

"Compare these three drafts against AAVE, UNI, and OP. Which has the worst insider concentration?" Claude calls the MCP, runs the math, returns a structured answer.

IN AGENT WORKFLOWS

Build a tokenomics agent that designs, validates, and audits in one loop. The MCP gives it real data + working math. No re-implementation.

How it feels

A worked example, end-to-end.

Watch one real exchange play through the four stages of an MCP call. Click a scenario to lock it, or let the demo cycle.

01 · FOUNDER

ask in plain english

“How does my draft compare to the real Uniswap launch on team & investor allocation?”

02 · CLAUDE

picks a tool, calls the mcp

tokenomics:compare_to_peers
{ "design": <my-draft>, "peers": ["uniswap"] }

03 · MCP

returns structured data

{ "team": { "you": 22, "uniswap": 21.5, "delta_pp": 0.5 },
  "investor": { "you": 30, "uniswap": 17.8, "delta_pp": 12.2 } }

04 · FOUNDER

reads the answer

Your team % is in line with Uniswap (22 vs 21.5). Investor share is 12pp heavier — that's your flag, not the team line.

AUTO-CYCLE

Install

Add the server to your AI assistant.

The endpoint is a remote MCP — no local install, no npm package. Just point your AI assistant at the URL.

Quick connect on Claude.ai

Copy the URL, paste it in Connectors.

On claude.ai go to Settings → Connectors → Add Custom Connector. Paste the URL. No auth required for the anonymous tier.

https://tokeneconomics.io/api/mcp

Same URL works in Cursor, Antigravity, Cline, Continue, and any MCP-aware client.

Or paste the config snippet into a desktop client

CLAUDE DESKTOP

Open your config: ~/Library/Application Support/Claude/claude_desktop_config.json (mac) or %APPDATA%/Claude/claude_desktop_config.json (Windows).

Add this entry to mcpServers:

{
  "mcpServers": {
    "tokenomics": {
      "url": "https://tokeneconomics.io/api/mcp"
    }
  }
}

CURSOR

Open ~/.cursor/mcp.json (or the workspace-level .cursor/mcp.json).

Add the server:

{
  "mcpServers": {
    "tokenomics": {
      "url": "https://tokeneconomics.io/api/mcp"
    }
  }
}

CLAUDE.AI / CONNECTORS

Settings → Connectors → Add Custom Connector. Set the URL to the endpoint below. No authentication required.

https://tokeneconomics.io/api/mcp

ANTIGRAVITY / CLINE / CONTINUE

These IDE-side AI tools all support remote MCP. Add a new server with the URL, no auth. Restart the agent for tools to register.

https://tokeneconomics.io/api/mcp

After install, your assistant will see 25 tools prefixed tokenomics:. Try “use the tokenomics MCP to compare PENGU and UNI on insider %”.

Tools

25 tools, 313 encoded projects.

Read tools return structured JSON. Compute tools wrap the same math the visual editor uses, so AI-generated designs and human designs are graded on the same scale.

01
list_real_projects
List 300+ real-world tokenomics with metadata + filtering.
02
get_project_tokenomics
Full tokenomics for one project (allocations, vesting, liquidity, descriptions).
03
get_industry_benchmarks
Median + p90 values for team %, investor %, treasury %, TGE circulating, etc.
04
compute_health_score
0–100 score with red/amber flags and remediation suggestions.
05
compare_to_peers
Per-metric ranking of any design vs all 300+ real protocols.
06
compute_sell_pressure
Month-by-month token unlocks + dollar projection over a horizon.
07
compute_tge_circulating
Exact TGE circulating supply, breakdown per allocation.
08
propose_design_from_brief
One-paragraph project brief → starting Design modelled on the closest comparable.
09
simulate_scenario
What-if sell-pressure projection at a multiplied launch price (price drops 50%, doubles, etc.).
10
find_similar_projects
Top-N closest comparables in the dataset by allocation shape + chain match.
11
audit_design
Risk audit beyond validation — flags insider-heavy splits, missing cliffs, unlocked LP, etc.
12
validate_design
Sanity-check sums, vesting math, integrity. Returns errors list.
13
compute_dilution
Each bucket's share of float at TGE / 12 / 24 / 36 / 60mo.
14
simulate_price_bands
Monte Carlo price bands — median + percentiles, drawdown & 2× odds.
15
analyze_sensitivity
Tornado — which demand assumption moves the price most.
16
get_unlock_schedule
Month-by-month circulating % + tokens unlocked.
17
model_demand
Token-utility config (staking / burn / buyback) → demand assumptions.
18
project_holder_concentration
Top-holder share of float projected over time.
19
diff_designs
Two candidate designs scored head-to-head.
20
search_by_metrics
Filter the 300+ by computed insider % / TGE float ranges.
21
explain_term
Glossary lookup — TGE, FDV, cliff, HHI, and more.
22
list_categories
Valid filter values — categories, chains, token types, with counts.
23
generate_narrative
AI · tagline + investor & community story from a design.
24
adversarial_review
AI · hostile-VC, CT-skeptic & CEX-committee objections.
25
get_upcoming_unlocks
Real tokens with significant unlocks coming up, sorted.

Test it without an AI

Hit the endpoint directly.

The server speaks JSON-RPC over HTTP. You can call it with curl to verify it’s alive before wiring up an AI client.

List available tools

curl -X POST https://tokeneconomics.io/api/mcp \
  -H 'Content-Type: application/json' \
  -H 'Accept: application/json, text/event-stream' \
  -d '{
    "jsonrpc": "2.0",
    "id": 1,
    "method": "tools/list",
    "params": {}
  }'

Built by 3UILD

We built Token Economics as a free tokenomics designer. We’re shipping the MCP because the next generation of token founders won’t open five tabs of calculators — they’ll ask Claude. We want to be in that conversation.

Talk to 3UILD→

Prefer a UI?

The same data + math powers a chat-first visual editor at Token Economics. Drafty drafts your design from a sentence; you refine in chat or by dragging the canvas directly.

Open the studio→

Find us on

Search for tokenomics-layer on any MCP marketplace.

Smithery↗mcp.so↗MCP Registry↗npm↗

The server is reachable at https://tokeneconomics.io/api/mcp regardless of which directory you discovered it through.