mcpField
I
Indraft

Indraft

Remote
indraft.ioBusiness & Marketing0Updated 8w ago

About Indraft

Indraft is a CRM for AI agents, with companies, contacts, opportunities, tasks, and interactions available without schema setup. It tracks the source of each value and every change, preventing weaker claims from silently replacing confirmed information.

Key features

  1. 01Manage companies, contacts, and opportunities
  2. 02Track tasks and interactions
  3. 03Record value provenance and changes
  4. 04Expose 45 tools according to credentials
  5. 05Support custom fields and object types

Use cases

  1. 01Let an agent read and update CRM records
  2. 02Review the sources and changes behind agent writes
  3. 03Extend the CRM structure for business needs

MCP support

Indraft provides an MCP server for reading and writing CRM data through clients that support Streamable HTTP and OAuth 2.1.

Configuration

Remote servers need no local install. Pick your client, then copy the config or install in one click.

Endpointhttps://api.indraft.io/mcp
Copy install command
Claude CodeTerminal
Cursor.cursor/mcp.json
VS Code.vscode/mcp.json
Claude Desktopclaude_desktop_config.json
Windsurf~/.codeium/windsurf/mcp_config.json

Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.

All endpoints
Streamable HTTPhttps://api.indraft.io/mcpAuth required

Tools

Details

From the project README

Indraft

A CRM your AI agent can actually run. Connect Claude, ChatGPT, Cursor or any MCP client and it can read and write companies, contacts, opportunities, tasks and interactions from the first request, with no schema to design first.

MCP server: https://api.indraft.io/mcp · indraft.io · Docs


Connect it

Claude
claude mcp add --transport http indraft https://api.indraft.io/mcp

Claude opens a browser to authorize. Nothing to paste, no API key to store.

In Claude Desktop or claude.ai, add it as a custom connector with the same URL.

ChatGPT

Settings → Connectors → Add custom connector, URL https://api.indraft.io/mcp. It authorizes over OAuth.

Cursor, Windsurf, VS Code, and other MCP clients

Any client that speaks Streamable HTTP and OAuth 2.1 works. Point it at https://api.indraft.io/mcp.

Then paste this and your agent will set itself up:

Set up Indraft for my business. Read its CRM schema first, then tell me what you think we do and how you plan to model it, and wait for my answer before you write anything.

It reads the schema before it writes, so it learns the object types, the limits and what it is allowed to do without you explaining any of it.


Why this exists

Most CRMs were built for a person clicking through forms, and an agent using one is doing screen-scraping with extra steps: guessing field names, inventing records, and overwriting things somebody already checked.

Indraft is built the other way round.

Every value records who asserted it and how. Observed, inferred, imported, supplied by a person, or confirmed by a person. Those rank, and a weaker claim cannot quietly overwrite a stronger one, so your agent's guess never overwrites what you checked. A write that was held says so rather than reporting success.

Every change is on the record. The actor, the surface it arrived through, and the field-level diff. Most changes can be undone.

It is populated from the first request. Companies, contacts, opportunities, tasks and interactions exist before you configure anything, with a pipeline that can actually close a deal. Custom fields and your own object types are there when you need them, defined by asking.

It refuses rather than guessing. A total that crosses currencies is refused unless you have stated a rate, and every converted answer names the rate it used and who asserted it. There is no bulk delete.


The tools

45, covering reading, writing, schema changes, merges, bulk imports and aggregation. An agent is only shown the ones its credential can actually call.

Full reference, generated from the server's own catalogue: TOOLS.md

Every tool carries a human-readable title and the applicable readOnlyHint, destructiveHint and openWorldHint, so a client can tell a read from a write before it calls one.


TypeScript SDK

For your own service, when you want the REST API rather than MCP. The source is in sdk/; an npm release follows.

import { IndraftClient } from "./sdk";

const indraft = new IndraftClient({ token: process.env.INDRAFT_TOKEN });

const { record } = await indraft.upsertCompany({
  name: "Halden Freight",
  domain: "halden.example",
  industry: "Logistics",
  // Say how you know. An import is weaker than a person, and Indraft enforces
  // that rather than trusting the last writer.
  assertionKind: "imported",
  idempotencyKey: "nightly-sync-halden",
});

Errors are typed and carry a recovery hint rather than a status code alone:

import { IndraftError } from "./sdk";

try {
  await indraft.upsertCompany({ name: "Halden Freight", industry: "Retail", overrideStanding: true });
} catch (error) {
  if (error instanceof IndraftError && error.code === "standing_override_refused") {
    // This workspace does not permit overrideStanding. Send the write again
    // without it: it applies to every field it is strong enough to move and
    // reports the rest in `held[]`, and a person can confirm a contested value.
    console.error(error.recoveryHint);
  }
}

A cross-currency total is not an error, it is a different answer. Ask for a sum over deals in several currencies with no rate on file and the aggregate comes back grouped per currency rather than summed, because a total that silently mixes currencies is the failure this product refuses to commit.

See sdk/ for the source. MIT.


Also available

  • REST API with a published OpenAPI description
  • Official MCP registry: io.indraft/mcp
  • Export everything, any time: one spreadsheet per table, the whole workspace as JSON, and a README explaining how the files join up. Nothing is left out and nothing is trimmed to fit.

Status

Indraft is live at indraft.io. Free tier, no card.

Issues and questions are welcome here. Security reports go to [email protected] rather than a public issue.

License

MIT for the SDK and everything in this repository. The hosted service has its own terms and privacy policy.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.