How to Connect Postman to Gemini via MCP

💡 No install? Use cloud.anythingmcp.com directly. Sign in, click Connectors → Postman, paste your credentials, mint an MCP API key and you're done. No Docker, no git clone, no local server.

Postman + Gemini

Work with Postman from your AI agent: browse workspaces, collections, folders and requests, environments, Spec Hub specifications and APIs, mock servers with their call logs, monitors and team users, search across them, and create or change collections, requests, environments, mocks, monitors and specs after you confirm.

Prerequisites

See the full setup instructions baked into the connector (visible in the in-app store when you select the connector). The required environment variables for this connector are:

POSTMAN_API_URL, POSTMAN_API_KEY

Step 1 — Get credentials

Setting up takes about 2 minutes. You need a Postman account with access to the workspaces.

1. In Postman open your avatar, Settings, API keys (https://go.postman.co/settings/me/api-keys) and click Generate API Key. Name it and copy the key; it starts with PMAK-.
2. Paste it into POSTMAN_API_KEY. It is sent in the X-Api-Key header and acts as your Postman user: the AI sees what you see in every workspace.
3. Set POSTMAN_API_URL to https://api.postman.com, or to https://api.eu.postman.com if your team is on an Enterprise plan in the Postman EU data region.
4. Install the connector and run postman_get_me: it shows your user id, team and the monthly API usage, and confirms the key.

Ids and uids: most objects have an id and a uid (<owner user id>-<id>). Mocks, monitors and postman_update_collection take the collection uid; postman_create_collection_request, postman_update_collection_request and postman_create_collection_folder take the plain collection id. Other reads accept either.

What it reads:

• You and your team: postman_get_me, postman_list_workspaces (with createdBy set to your user id for "my workspaces"), postman_get_workspace (the collections, environments, mocks, monitors and APIs it holds), postman_list_team_users and postman_search, which finds requests, collections, workspaces, environments, documents, flows or specs by text (your team only by default, public ones with ownership).
• Collections in Postman Collection Format v2.1: postman_list_collections, postman_get_collection (model=minimal first for a big one), postman_get_collection_folder and postman_get_collection_request with method, URL, headers, body, auth and scripts.
• Environments: postman_list_environments and postman_get_environment with key, value, type and enabled flag of every variable.
• API design: Spec Hub with postman_list_specs, postman_get_spec, postman_list_spec_files, postman_get_spec_file and postman_get_spec_definition (the whole resolved OpenAPI or AsyncAPI definition); older API Builder APIs with postman_list_apis and postman_get_api.
• Mocks: postman_list_mocks, postman_get_mock and postman_get_mock_call_logs, what clients sent to the mock and what it answered (at most 100 logs or 6.5 MB per call; headers and bodies only with include).
• Monitors: postman_list_monitors and postman_get_monitor, whose lastRun shows the latest status, times and passed and failed assertions.

What it writes, after you confirm: postman_create_collection, postman_update_collection (name, description, variables, auth; folders and requests stay as they are), postman_create_collection_folder, postman_create_collection_request, postman_update_collection_request (the fields passed replace the old ones), postman_create_environment, postman_update_environment, postman_create_mock, postman_create_monitor, postman_run_monitor, postman_create_spec (OpenAPI, AsyncAPI, protobuf, GraphQL, Smithy) and postman_update_spec_file. The AI says what it will create or change and in which workspace, and calls after your yes. Everything written is visible to the workspace's members. Keep in mind:

• postman_run_monitor sends the collection's real requests to their APIs and uses a monitoring run of your plan; postman_create_monitor keeps doing so on its schedule.
• A new mock is public unless private is true.
• postman_update_environment takes JSON Patch operations of one kind per call; the AI reads the environment first to get the variable positions (values[0] is /values/0).
• The Postman API cannot run a collection directly. Use a monitor, or the Postman CLI in your own CI.

All 35 tools are active after installation; none is switched off, and no tool deletes a workspace, collection, request, environment, mock, monitor or spec.

Secrets: environment variables of type secret come back with their values when they are shared, and collection variables or request auth may hold tokens. The AI does not repeat such values unless you ask.

Data formats: collections and requests use Postman Collection Format v2.1 (item holds folders and requests). Dates are ISO 8601 in UTC.

Paging: postman_list_workspaces, postman_list_apis, postman_list_specs, postman_list_monitors, postman_get_mock_call_logs and postman_search return meta.nextCursor; the AI calls again with cursor until it is empty. postman_list_collections pages with limit and offset; its name filter matches the exact name and cannot be combined with them.

Limits: Postman allows 300 requests per minute per user; postman_list_collections, postman_list_workspaces, postman_get_workspace and the monitor endpoints only 10 calls in 10 seconds. Each plan also has a monthly quota of API calls, which postman_get_me shows under operations. A 429 with rateLimited means wait (Retry-After gives the seconds); serviceLimitExhausted means the plan's quota is used up and a team admin has to act.

Common errors: 401 means the key is wrong, expired or revoked, or belongs to the other data region (check POSTMAN_API_URL). 403 means your role in that workspace does not allow the call. 404 means the id does not exist or you cannot see it. 400 names the wrong field, for example a collection id where a uid is needed.

The connector does not delete anything, and does not handle forks and pull requests, comments, tags, governance, the API catalog, the private API network, SCIM, audit logs, billing or the older API Builder writes.

Step 2 — Install the adapter

curl -fsSL https://raw.githubusercontent.com/HelpCode-ai/anythingmcp/main/docker-compose.quickstart.yml -o docker-compose.yml
printf 'JWT_SECRET=%s\nENCRYPTION_KEY=%s\n' "$(openssl rand -hex 32)" "$(openssl rand -hex 32)" > .env
docker compose up -d

Step 3 — Add the connector in Gemini

Your server URL: in AnythingMCP, open MCP Servers → the server this connector is on, and copy its URL (https://cloud.anythingmcp.com/mcp/…). Use it wherever this guide shows YOUR_SERVER_ID.

Gemini CLI reads MCP servers from ~/.gemini/settings.json (on Windows %USERPROFILE%\.gemini\settings.json). Add:

{
  "mcpServers": {
    "anythingmcp": {
      "httpUrl": "https://cloud.anythingmcp.com/mcp/YOUR_SERVER_ID",
      "headers": { "X-API-Key": "YOUR_MCP_API_KEY" }
    }
  }
}

1. Create an MCP API key in AnythingMCP under MCP Servers → your server → API keys, and paste it in place of YOUR_MCP_API_KEY.
2. Save the file and restart gemini.
3. Run /mcp inside Gemini CLI (or gemini mcp list in your shell): anythingmcp should show as connected.
4. Start prompting: Gemini CLI can now call the Postman tools.

Available tools

Example prompts

• "Which collections are in our Payments workspace, and which requests does the Checkout collection contain?"
• "Did the nightly monitor for the orders API pass its last run? How many assertions failed?"
• "Show me the last 20 calls to our checkout mock server that got a 404."
• "Add a POST request 'Create refund' to the Payments collection with a JSON body for orderId and amount."
• "Give me the resolved OpenAPI definition of the Orders spec and list its endpoints."

FAQ

Where does Gemini CLI look for MCP servers? In ~/.gemini/settings.json for your user, or in .gemini/settings.json inside a project. gemini mcp list shows each server and whether it connects.

Next steps

• Connector landing page
• Connect to Claude
• Connect to ChatGPT
• Connect to Cursor
• Connect to Meta Muse
• Connect to Microsoft Copilot