The MCP control plane
Governed MCP servers for your whole company
Connect every product your teams use, compose curated toolboxes, and hand each person a personal, permission-clamped MCP endpoint for Claude, Cursor, or any MCP client. No shared API keys. No shadow servers. Full audit trail.
{
"mcpServers": {
"elaichi-hr-toolbox": {
"url": "https://api.elaichi.ai/mcp/elch_••••••••••••"
}
}
}
// One personal endpoint per user — clamped by roles, restrictions & frozen params.
The problem
AI access to your SaaS stack is a free-for-all
Everyone is pasting personal API keys into ad-hoc MCP servers. Nobody knows who can reach what, credentials sprawl across laptops, and there is no record of what an agent actually did. Elaichi replaces that with one governed control plane: connections are vaulted once, access is scoped by role, and every tool call is logged.
How it works
Connect → Compose → Mint → Govern
01
Connect
Pick from 200+ connectors and finish a hosted connect flow. Credentials go straight into an encrypted vault — your team never handles API keys.
02
Compose
Every connection becomes a toolbox of MCP tools instantly. Curate template toolboxes: rename tools, freeze parameters, set defaults, share with people, teams, or the whole org.
03
Mint
Create an MCP endpoint on any toolbox and paste it into Claude Desktop, Cursor, or Dust. Rotate tokens anytime; large toolboxes auto-compact to protect the context window.
04
Govern
Roles decide who can do what. Restrictions decide which connectors and tools are even visible. Every call lands in an append-only audit log.
Capabilities
Built agent-first, governed by default
200+ connectors, zero authoring
Every documented method of every connector is already an LLM-callable tool. Connect an account and the toolbox is just there.
Delegated connections
Share a toolbox that rides on your connection. Teammates call the API through it — they never see, hold, or leak a credential.
Frozen parameters
Pin tool arguments to fixed values stripped from the schema the model sees. The AI physically cannot override them.
Synthetic tools
Compose steps across multiple connections into one composite tool — executed server-side through the same governance pipeline.
Context-blowup protection
Above ~30 tools, endpoints advertise just search_tools and execute_tool, keeping big toolboxes out of the model’s context window.
Three-layer governance
RBAC roles, per-resource ACLs, and connector/tool restrictions — enforced at connect, advertise, and execute time.
Audit everything
An append-only log covers every privileged action, including each MCP tool call with status and duration. Forward it to your Datadog.
Custom & forked connectors
Author a connector from a JSON config or fork any public one — with lineage tracking and selective pull-from-upstream.
Enterprise-ready identity
SAML/OIDC SSO with enforced domains, SCIM v2 provisioning, group-to-role mappings, TOTP MFA, and BYOK encryption.
Want the full tour? Explore the product →
Desktop companion
Nobody hand-edits JSON
The Elaichi menubar app lists your toolboxes, mints MCP servers, and writes claude_desktop_config.json and Cursor’s mcp.json for you — so rolling out AI tooling doesn’t require a config tutorial for every teammate.
- ✓ One-click setup for Claude Desktop and Cursor
- ✓ Tokens encrypted with the macOS Keychain
- ✓ See and manage every MCP server you hold
Security
Credentials the AI never sees
Secrets live in a dedicated vault, encrypted at rest with AES-256-GCM. Tokens are hashed and shown once. Enterprise plans bring their own AWS KMS keys. Tools execute server-side — neither users nor models ever touch a credential.
FAQ
Frequently asked questions
What is an MCP control plane?
An MCP control plane is a central system that manages how a company creates, shares, and governs Model Context Protocol (MCP) servers. Instead of each employee minting ad-hoc MCP servers with personal API keys, Elaichi connects your SaaS tools once, turns them into curated toolboxes, and hands every user a personal MCP endpoint clamped by roles, restrictions, and audit logging.
How does Elaichi work?
Four steps. Connect — pick from 200+ connectors and finish a hosted connect flow; credentials are vaulted and encrypted. Compose — every connection instantly becomes a toolbox of MCP tools, or you curate a template toolbox with renamed tools, frozen parameters, and defaults. Mint — create an MCP endpoint on any toolbox and paste the URL into Claude Desktop, Cursor, or any MCP client. Govern — roles decide who can do what, restrictions decide which tools are even visible, and every call lands in an append-only audit log.
How do I give my team MCP servers without sharing API keys?
With delegated connections. In Elaichi, a shared toolbox can expose the sharer's connection: recipients call the third-party API through it without ever seeing credentials. Every call is clamped server-side by the toolbox's tool list, frozen parameters, and restrictions.
Which AI clients work with Elaichi?
Any MCP client that speaks Streamable HTTP — Claude Desktop, Cursor, and Dust connect directly today. Elaichi MCP endpoints speak standard MCP JSON-RPC (initialize, tools/list, tools/call), and the desktop companion app writes Claude Desktop and Cursor config files for you.
What are frozen parameters?
Frozen parameters are tool arguments fixed by the toolbox curator and stripped from the schema the model sees. They are force-merged at execution time, so the AI can never see or override them — for example pinning a Google Drive tool to one specific folder.
What happens when a toolbox has too many tools for the model's context?
Toolboxes exposing more than about 30 tools automatically advertise just two meta-tools — search_tools(query) and execute_tool(name, arguments) — so large toolboxes stay out of the model's context window. This behavior is configurable per MCP server.
Does Elaichi store my SaaS credentials?
Third-party credentials are held in a dedicated credential vault, encrypted at rest with AES-256-GCM, and are never exposed to end users or to the AI. Session, API, and MCP tokens are stored as keyed hashes and shown only once. Enterprise plans can bring their own AWS KMS keys.
Can I build a connector for an internal API?
Yes. Custom connectors let you author a connector from a JSON config — base URL, auth, resources, and methods — or fork any of the 200+ public connectors, including its documentation. Forks track upstream lineage, so you can review and selectively pull upstream improvements later.
How is Elaichi different from running my own MCP servers?
Self-hosted MCP servers multiply credentials, have no shared governance, and leave no audit trail. Elaichi centralizes connections with vaulted credentials, applies role-based access control and allow/blocklists at connect, advertise, and execute time, logs every tool call, and lets admins revoke or rotate any endpoint instantly.
Give every teammate a governed MCP server
Start a 14-day Gold trial — connect a product, curate a toolbox, and paste an endpoint into Claude or Cursor in minutes.