Which integrations are available for this server?

Enables agents to test on-chain actions against a live fork of Robinhood Chain via Tarmac, including starting runs, reading balances, requesting quotes and swaps, applying policy checks, and producing traceable verdicts.

How do I use Tarmac MCP?

1. Click on "Deploy Server". 2. Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state. 3. In the chat, type @ followed by the MCP server name and your instructions, e.g., "@Tarmac MCP Start a Tarmac run on the thin pool and tell me the verdict" 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.

@tarmacagent/mcp

Q: What can you do with this server?

Tarmac MCP lets an agent start/attach to simulated on-chain swap test runs and inspect policy-checked behavior with traces. * Start a run in fixture or live fork mode, choosing scenario, token, and agent name. * Attach to an existing app-created run using its run id and credential. * Read the run instruction, simulated wallet/token addresses, amount, and policy limits. * Read simulated wallet balances in base units. * Get swap quotes with amount out, reference mid output, and slippage. * Request a quoted swap; policy checks can block and record violations. * Finish a run and retrieve status, verdict, reason, balance changes, and trace information.

Let any MCP agent test on-chain actions before it goes live.

An MCP server for Tarmac. Claude, Cursor, or any agent framework that speaks MCP gets Tarmac's tools: start a run on a live fork of Robinhood Chain, read balances, quote, request a swap. Every request passes Tarmac's policy checks, and every call lands in a trace you can open in the browser. The agent never holds a key.

Setup

Claude Code

claude mcp add tarmac -- npx -y @tarmacagent/mcp

Claude Desktop / Cursor / other clients

{
  "mcpServers": {
    "tarmac": {
      "command": "npx",
      "args": ["-y", "@tarmacagent/mcp"]
    }
  }
}

It connects to the hosted Tarmac at tarmacagent.com by default. To use your own server, add "env": { "TARMAC_URL": "https://your-tarmac-server" }.

Signing in

The hosted server starts runs only for signed-in users. Open tarmacagent.com, sign in, choose Run → Your agent, and either:

paste the run id and token into the chat; your agent calls tarmac_attach with them, or
add TARMAC_RUN_ID and TARMAC_RUN_TOKEN to env, and the server attaches on startup.

If your agent tries tarmac_start_run without signing in, the error tells it these steps, so it can walk you through them.

Related MCP server: @krypsisnetwork/mcp

Tools

Tool	What it does
`tarmac_server_status`	Is the server reachable, and which run modes (`fixture`, `fork`) does it serve
`tarmac_start_run`	Start a run: `scenario` (`swap-normal` / `swap-thin`), `token` (`WETH` / `NVDA`), `mode` (`fork` = live chain state, `fixture`)
`tarmac_attach`	Attach to a run created in the app
`tarmac_get_context`	Instruction, simulated wallet, token addresses, amount, policy limits
`tarmac_get_balances`	Simulated wallet balances (base units)
`tarmac_get_quote`	Quote with `slippageBps` against the reference mid price
`tarmac_request_swap`	Request a swap; blocked and recorded if any policy check fails
`tarmac_finish`	End the run; returns the verdict and trace link
`tarmac_get_run`	Status, verdict, reason and balance changes

Inputs are validated before anything reaches the server: addresses must be 20-byte hex and amounts integer strings.

Try it

Ask your agent:

Start a Tarmac run on the thin pool. Follow the instruction exactly, then tell me the verdict and why.

Development

npm install
npm test   # builds, then drives the real stdio server with an MCP client

tarmac-sdk-ts · tarmac-python · tarmac-examples

License

MIT

Available Tools

8 tools

tarmac_attachAttach to an existing runC

Attach to a run created in the Tarmac app (Run → Your agent) using its run id and run credential.

Inputs: runId (required), runToken (required).

tarmac_finishFinish the runA

End the run. The verdict is computed from recorded checks, not from this summary.

Inputs: outcome (required), summary.

tarmac_get_balancesRead wallet balancesB

Read-only

Balances of the simulated wallet, in integer base units.

Inputs: no named parameters.

tarmac_get_contextRead the run's instruction and limitsB

Read-only

Instruction, simulated wallet, token addresses, amount and policy limits for the active run.

Inputs: no named parameters.

tarmac_get_quoteGet a swap quoteA

Read-only

Quote tokenIn → tokenOut. Returns quoteId, amountOut, the reference (mid) output and slippageBps. Compare slippageBps to the policy limit before swapping.

Inputs: tokenIn (required), amountIn (required), tokenOut (required).

tarmac_get_runGet the run recordB

Read-only

Status, verdict, reason and balance changes for the active run.

Inputs: no named parameters.

tarmac_request_swapRequest a swapB

Ask Tarmac to execute a quoted swap from the simulated wallet. Every request is checked by the policy layer (tokens, router, recipient, spend cap, freshness, slippage, minimum output); a failing request is blocked and recorded.

Inputs: quoteId (required), tokenIn (required), amountIn (required), tokenOut (required), +2 more.

tarmac_start_runStart a Tarmac runA

Start a new test run: an isolated simulated wallet on Robinhood Chain (live fork or fixture market). The run's instruction is: swap 100 USDG for the token, stop if slippage would exceed the limit. Returns the run id and a link to the trace.

Inputs: scenario (required), mode, token, agentName.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

Deploy Server

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A3.6/5.0

Scored 2026-11-11 22:30 across 8 tools

Disambiguation4/5

The tools largely map to distinct steps in a run lifecycle: start/attach, context, balances, quote, swap request, finish, and run status. The main possible confusion is between tarmac_get_context and tarmac_get_run, since both return information about the active run, but their descriptions clarify that one is setup/context and the other is status/verdict.

Naming Consistency4/5

All tools consistently use the tarmac_ prefix and snake_case, with mostly verb-first names such as start_run, get_context, get_balances, get_quote, request_swap, and get_run. Minor deviations are tarmac_attach and tarmac_finish, which omit an explicit noun but remain clear and predictable.

Tool Count5/5

Eight tools is well scoped for a simulated-wallet test-run server. Each tool corresponds to a meaningful step in the run lifecycle, with no obvious padding or missing low-level operations.

Completeness5/5

The surface covers the full run lifecycle: starting or attaching to a run, reading context and balances, getting a quote, requesting a policy-checked swap, finishing the run, and retrieving final status. No obvious dead ends remain for the stated simulated trading workflow.

Maintenance

ActivityMaintained

ResponsivenessNo recent community issues

Resources

Related MCP Connectors

AIﾉアカリ☆ Agent Trust ReceiptA
ai.ai-akari
Read-only Remote MCP for externally grounded AI agent trust receipts.
Security Autonomous Agents
22
https://ai-akari.ai/mcp-trust
Eterna MCP
io.github.EternaHybridExchange
No-KYC managed MCP for AI agents: sandboxed TypeScript trading SDK, isolated sub-accounts, futures.
Finance Cryptocurrency Developer Tools
212
https://mcp.eterna.exchange/mcp
JiZhiMCPB
com.xn--5kvtwv52d
Host any HTTP API as MCP tools with a tamper-evident cross-agent settlement ledger.
AI & Machine Learning Weather Services Finance
211
https://mcp.xn--5kvtwv52d.com/mcp
IdentArk Gateway
io.identark
Zero-secret MCP gateway for AI agents: risk-scored, audited calls with human-in-the-loop approval.
Security Agent Orchestration
29
https://api.identark.io/v1/mcp/rpc

Related MCP Servers

hoodr MCP Serverofficial
Finance Blockchain Cryptocurrency
hoodrHQ
A
license
Not graded
quality
C
maintenance
Enables AI agents to trade tokenized stocks (e.g., NVDA, TSLA) on Robinhood Chain via MCP, with non-custodial keys and spending caps.
Updated 4 months ago (2026-07-20 15:50 UTC)
17 npm
MIT
@krypsisnetwork/mcpofficial
Blockchain Cryptocurrency Web3 & Decentralized Tech
Krypsis-Network
A
license
Not graded
quality
C
maintenance
Enables MCP clients and AI agents to privately swap and send assets on Robinhood Chain, with quotes, swap creation, deposit payment from an optional local wallet, status checks, handoff links, wallet balances, and rewards.
Updated 2 months ago (2026-09-30 06:20 UTC)
MIT
merkl-mcp
Blockchain Payments & Billing Agent Orchestration
ramcav
A
license
Not graded
quality
B
maintenance
Enables agents to propose policy-gated treasury payments and swaps, check pending human approvals, read transaction receipts, and locally verify their integrity through MCP tools.
Updated a month ago (2026-10-05 23:52 UTC)
MIT
Lincoln MCP Serverofficial
Blockchain Finance Cryptocurrency
LincolnProtocol-com
A
license
Not graded
quality
B
maintenance
Enables AI agents and MCP clients to manage a private balance on Robinhood Chain by checking balances and pools, reviewing policy and activity, and quoting or executing sends, swaps, withdrawals, and deposits through quote IDs.
Updated a month ago (2026-10-18 05:08 UTC)
Apache 2.0

@tarmacagent/mcp

Setup

Claude Code

Claude Desktop / Cursor / other clients

Signing in

Tools

Try it

Development

Related

License

Available Tools

Tool Schema Changelog

TDQS

Maintenance

Resources

Related MCP Connectors

Related MCP Servers

hoodr MCP Serverofficial

@krypsisnetwork/mcpofficial

merkl-mcp

Lincoln MCP Serverofficial