Boltz API

Predict biomolecular structure and binding, screen compound libraries, design binders.

Add to Agentman Studio

Opens Agent Studio, where connecting is one click. The connector URL below works in any MCP client.

Paste it into any MCP client. Setup docs

Resources

What the Boltz API connector does

The Boltz API MCP server connects Boltz — a biomolecular structure and binding prediction platform — to Claude, ChatGPT and any MCP-compatible agent. Anthropic's directory lists 10 tools that predict structures and screen compound libraries against a target. Three of them submit GPU jobs that cost real money, which is the first thing to know about this connector.

Verified connector

Listed by Anthropic as a partner connector in its Connectors Directory.

Connection checked by Agentman on .

Anthropic states this reflects the level of review a connector received, not a security audit.

Boltz API tools (10)

  • boltz_estimate_protein_screen
  • boltz_estimate_small_molecule_screen
  • boltz_estimate_structure_and_binding
  • boltz_get_guidance
  • boltz_get_job_status
  • boltz_get_screen_results
  • boltz_get_structure_and_binding_prediction
  • boltz_start_protein_screen
  • boltz_start_small_molecule_screen
  • boltz_start_structure_and_binding

Tool names from Anthropic's directory listing. This server requires sign-in, so we could not read tool descriptions or parameter schemas.

Limits

  • Three tools spend money and no scope separates them from the reads. Boltz advertises one application scope, compute:run, covering both estimating and submitting. The boundary between a free estimate and a billed run lives entirely in your client's approval prompts.
  • Boltz publishes no tool list for this server. The 10 names come from Anthropic's directory. We found no MCP enumeration in either Boltz documentation site or any public Boltz repository, so we could not diff them against a vendor list — none exists.
  • The directory description names capabilities the tool list does not carry. It says the server can design small molecules and protein binders; no tool among the ten starts a design job, though the REST API documents all three design types.
  • We could not read tool schemas or safety annotations. The endpoint returned 401 to an anonymous request, so no parameter-level detail is published here.
  • Cost estimates are approximate, and least precise where the bill is largest. Boltz states that final billing follows the exact compute measured at run time, and that for large library screens the estimate is extrapolated from a sample.
  • Results expire. Workspace data retention defaults to 7 days after a run terminates, with a 14-day ceiling and a 30-day hard stop on runs that never terminate. Download what you need.
  • Test mode is not available over this connector. Boltz's no-GPU, no-billing test keys are API keys; the MCP server authenticates with OAuth.
  • Rate and queue limits apply per organization and workspace, at 100 requests per 10 seconds and 5,000 pending predictions / 500 pending pipeline runs respectively.
  • Boltz's own docs never mention this connector. Its agent-integration guidance covers the boltz-api CLI and marketplace plugins instead, so a reader following the vendor's instructions installs something else.

Frequently asked questions

Does the Boltz MCP server spend money when an agent calls a tool?

Three of the ten tools do. The boltz_start_structure_and_binding, boltz_start_small_molecule_screen and boltz_start_protein_screen tools submit GPU jobs, and Boltz bills by compute measured at run time. The three matching estimate tools return a price without creating a run or using any GPU, so an estimate call is free and a start call is not.

What OAuth permissions does the Boltz MCP server request?

Five scopes, four of which are identity claims. Boltz advertised openid, profile, email, offline_access and compute:run on 2026-08-22. Only compute:run is an application scope, and it is the single grant that covers both reading a job's status and submitting a new billable one. No scope separates the read tools from the spending ones.

How do I cap what an agent can spend on Boltz?

Set a workspace spending limit, not an MCP setting. Boltz documents a hard lifetime cap per workspace, denominated in milli-USD, where 1000 means USD 1.00. Boltz reserves each job's estimated cost against the ledger before the work starts and rejects anything over the cap with the error workspace_spending_limit_reached. Setting the limit to zero blocks new billable work.

How much does a Boltz prediction cost?

Boltz's pricing page lists USD 0.025 per small molecule for the small-molecule pipeline, and a crop-size ladder for protein design and Sandbox running from USD 0.025 at 0-256 residues to USD 0.40 above 2048. The Professional plan carries no subscription and includes up to 200 free predictions each month. A library screen multiplies the per-unit figure by the library size.

Can the Boltz connector delete my prediction data?

Not with the ten tools Anthropic lists. Boltz's REST API has an irreversible delete-data endpoint on every prediction, design run and library screen, and no tool among the ten carries a delete verb or maps to it. Data still disappears on its own schedule: workspace retention defaults to seven days after a run reaches a terminal state, capped at fourteen.

Is the Boltz MCP server the same as the Boltz Claude Desktop extension?

No, they are two different servers with two different tool sets. The remote server at mcp.boltz.bio is what Anthropic's directory lists. Boltz also ships a local .mcpb desktop extension whose eleven tools share exactly one name with the remote listing, boltz_get_guidance, and which shells out to a boltz-api CLI you install yourself.

Can I test a Boltz agent integration without being billed?

Yes, with a test-mode API key, though not over the MCP connector. Boltz documents keys prefixed sk_bc_admin_test_ and sk_bc_ws_test_ that run the full request, poll and fetch lifecycle against real endpoints and return synthetic results with no GPU usage and no billing. The MCP server signs in with OAuth rather than an API key, so test mode is a CLI and SDK path.

What rate limits apply to the Boltz API?

One request limit and one queue cap. Boltz documents 100 requests per 10-second sliding window scoped per organization, returning HTTP 429 with Retry-After. Separately, queue caps reject new submissions at 5,000 pending predictions and 500 pending pipeline runs per organization, or 1,000 and 100 per workspace. Test-mode submissions are exempt from the queue caps.

Sources

More Healthcare connectors

  • AdisInsight

    Query Springer Nature's drug-pipeline and trial database from an AI agent.

  • AdvancedMD

    Look up patients, find open slots, book appointments, and work denials and remittances in AdvancedMD from your AI agent.

  • AllTrails

    Search AllTrails' trails by place, map bounds or name and pull details and a 7-day trailhead forecast from your AI…

Browse all MCP connectorsClaude and ChatGPT setup guideBrowse AI skills

Product names and logos belong to their owners. Agentman is not affiliated with a connector's maker unless the page says the connector is built and run by Agentman.

Use it in an agent

Put Boltz API to work.

Connect Boltz API once and your agents call these tools on their own: on a schedule, in a workflow, with nobody at the keyboard.

Open in Agentman Studio