Blog · Page 3
Governing AI agents in the apps you already run
How the gateways compare, what to restrict, what the audit log has to capture, and what each team does with it.
MCP tools not showing up? Start here
MCP tools not showing in Claude, ChatGPT or Cursor? With Elaichi, connected tools are never listed by design. Check these five things, in order.
Raajshekhar Rajan
Elaichi vs Composio: who writes your connectors?
Elaichi vs Composio: Elaichi writes most of its connectors and serves them on one company-wide MCP endpoint. Composio covers more apps, one endpoint per team.
Roopendra Talekar
Claude and ChatGPT connectors vs one MCP endpoint
Claude and ChatGPT connectors fit one client and a few apps. Once a second client repeats the setup, one governed MCP endpoint costs less to run.
Roopendra Talekar
Least privilege for AI agents without breakage
Least privilege for AI agents starts from what your pilot actually called: narrow to that recorded set, then confirm the rule landed before you trust it.
Raajshekhar Rajan
Engineers using ChatGPT at work: block or scope it
For engineers using ChatGPT at work, scope access rather than block it. A block moves the paste to a phone; scoped access removes the reason to paste.
Nachi Raman
SOC 2 evidence for AI agents: CC6 and CC7
SOC 2 evidence for AI agents maps to CC6.1, CC6.2, CC6.3 and CC7.2: who can reach what, how access starts and ends, and a log of each executed tool call.
Roopendra Talekar
Xero in Claude, without letting it edit invoices
Xero in Claude for a finance team: analysts read invoices and reports, no assistant can edit or void an invoice, and one named person drafts supplier bills.
Uday Gajavalli
Lunar MCPX alternative: count your servers first
With no MCP servers to front, the Lunar MCPX alternative is a hosted server like Elaichi. With several, a gateway like MCPX still fits.
Raajshekhar Rajan
Should marketing get HubSpot inside ChatGPT?
Yes: marketing can have HubSpot inside ChatGPT to read contacts and campaigns and draft work, with email sends and deletes restricted and the portal pinned.
Roopendra Talekar
Why blocks match tool names but allows don't
In Elaichi, blocks match tool names or the operation behind them, while allows match the operation only, so a renamed tool can never widen a role's reach.
Uday Gajavalli
Connect Elaichi to ChatGPT
To connect Elaichi to ChatGPT, a workspace admin adds one MCP endpoint as a custom app and each person signs in with OAuth. No API key, no per-user URL.
Raajshekhar Rajan
Each rep's own Salesforce access, in ChatGPT
Salesforce access in ChatGPT works per rep: each call runs on the rep's own connection, so Salesforce's sharing rules decide which accounts they see.
Nachi Raman
How MCP tool search picks one tool from hundreds
MCP tool search in Elaichi scores tools by matching words, then returns nothing unless a tool covers at least half the query, with rare words weighted most.
Raajshekhar Rajan
Running your own MCP servers: the real cost
Running your own MCP servers wins for one internal tool. Managed wins once credentials, token refresh, per-user sign-in and audit multiply.
Roopendra Talekar
When you don't need an MCP gateway yet
Per-user connectors plus SSO are enough while you can name everyone connected. That is when you don't need an MCP gateway, and four signals end it.
Roopendra Talekar
Lock AI agent tool arguments, like a wire's payee
Lock AI agent tool arguments with frozen parameters: Elaichi removes the field from the model's schema and writes your value over whatever the call sends.
Uday Gajavalli
Ironclad contracts in ChatGPT, signing blocked
Legal reads Ironclad contracts in ChatGPT through Elaichi. The connector has no signing or approval tool, and a role rule blocks launching or cancelling.
Nachi Raman
Employees pasting customer data into AI
No single log records employees pasting customer data into AI. Measure it from three partial sources, report a range, and remove the reason to paste.
Roopendra Talekar
Zapier MCP alternative for a whole company
Elaichi is a Zapier MCP alternative: one organization endpoint with role restrictions and an audit log. What each costs, and when Zapier fits.
Roopendra Talekar
CASB and AI agents: what each layer can see
CASB and AI agents: a CASB sees who reached which AI service and how much data moved, but not which tool an agent ran or which account it changed.
Raajshekhar Rajan
Connect Elaichi to Cursor for a whole team
To connect Elaichi to Cursor, add one URL to mcp.json or share it as a Team MCP server, and each developer signs in with OAuth. No API key.
Nachi Raman
SCIM and AI agents: where provisioning stops
SCIM and AI agents meet at the user account: SCIM creates, updates and deactivates it, but it never decides which tools an agent may call.
Uday Gajavalli
MCP endpoint per team, per person or per company?
One endpoint per company usually beats an MCP endpoint per team or a server per person: each AI client needs one entry, and a leaver is one action.
Roopendra Talekar
The MCP context window problem, and a fix
The MCP context window problem: each listed tool sits in the model's prompt on every turn. Elaichi lists no connected tools; the model searches for them.
Uday Gajavalli
Put agents to work on your own systems
14 days on Gold, no credit card. Start with one app and one team.
- Works with
-
and any other MCP client, or the Elaichi Agent.
- When the trial ends
- Nothing is deleted. Connections, roles and the audit log stay where they are, so subscribing picks up exactly where you left off.