Read AI-gateway analytics, configs, virtual keys, workspaces and users; log request feedback. Hosted by usefulapi: connect it to Claude, ChatGPT, Cursor or any MCP client.
Turn on the Portkey connector, then ask in plain words. The AI picks the tools.
For AI platform engineers who send LLM requests through the Portkey gateway.
The AI reads a time series for one metric, such as cost, requests, tokens, latency, error rate or cache hit rate. It can scope by virtual keys, configs or metadata.
The AI reads analytics grouped by models, users or metadata over a time window. You see which model or user drives the most usage.
The AI lists configs (routing, fallback, retry and cache rules) and reads one by slug, including its routing rules.
The AI lists virtual keys with their provider and limits, and API keys as metadata only. Secrets are never returned.
The AI lists workspace members with roles, organization users and pending invites, and reads the audit log of admin actions.
The AI writes a feedback record on a request by trace id, with a score, an optional weight and metadata.
Changes dataThe login page asks for your Portkey API key.
Find it in the Portkey app → API Keys (an organization-level key is needed for admin/analytics endpoints). Your key is used only to call the API on your behalf and to derive a private account id; it is never shown to anyone.
Team and Enterprise: an Owner adds the connector for the organization first.
No Add custom MCP server option? It depends on your account or workspace settings: see OpenAI's guide.
Run this command, then run /mcp in Claude Code to log in:
Add to Cursor or add this to ~/.cursor/mcp.json:
{
"mcpServers": {
"portkey": {
"url": "https://portkey.usefulapi.io/mcp"
}
}
} Add to VS Code or add this to .vscode/mcp.json:
{
"servers": {
"portkey": {
"type": "http",
"url": "https://portkey.usefulapi.io/mcp"
}
}
} Run these commands. The second one opens the login page:
Windsurf, Cline, Zed and other MCP clients: add this URL as a remote MCP server (Streamable HTTP). The client then opens the login page in your browser.
Add only the URL. Do not add an Authorization header or an API key to the client config. The server signs you in with OAuth: the login page asks for your Portkey API key. If the config has such a header, remove it: some clients then send that header instead of the login token, and every call fails with 401.
| Plan | Price | Limit |
|---|---|---|
| Free | $0 | 50 tool calls / month |
| Proper user | $9/mo · $90/yr | 10,000 tool calls / month |
| Pro+per user | $39/mo · $390/yr | 100,000 tool calls / month |
Each paid plan covers this Portkey server only. Ask your AI to run portkey_upgrade and choose Pro or Pro+, billed monthly or yearly (it returns a Stripe Checkout link). If you already have Pro, ask to change it to Pro+. The change requires your confirmation, keeps your billing interval and charges the price difference for the rest of the period immediately. Cancel any time with portkey_cancel_subscription: your paid plan continues to the end of the paid period, with no refund for the current period, and asking to resume the same plan with portkey_upgrade before then undoes the cancel. Or write to [email protected].
portkey_get_analytics_graphportkey_get_analytics_groupportkey_list_configsportkey_get_configportkey_list_api_keysportkey_get_api_keyportkey_list_virtual_keysportkey_get_virtual_keyportkey_list_workspacesportkey_get_workspaceportkey_list_workspace_membersportkey_list_usersportkey_get_userportkey_list_user_invitesportkey_list_audit_logsportkey_create_feedbackportkey_usage_statusportkey_request_featureportkey_upgradeportkey_cancel_subscriptionTrack cost, tokens, errors and latency over time. Compare models, users or metadata by usage. Review gateway configs and their routing rules. Check virtual keys, their limits and API keys. Review users, invites and admin actions. Log a quality score for a request. See what you can do for the details.
No. usefulapi is an independent service. It is not affiliated with or endorsed by Portkey. The server calls the Portkey API with your own Portkey access, so it sees only the data that your account can see.
Yes. In Claude, add this server's URL as a custom connector. In ChatGPT, add it as a custom MCP server in Plugins. Then log in once. Cursor, VS Code, Claude Code and Codex work too. See the setup steps.
The login page asks for your Portkey API key. See Before you connect for where to find it.
No. Add only the URL. Do not add an Authorization header or an API key to the client config. The server signs you in with OAuth: the login page asks for your Portkey API key. If the config has such a header, remove it: some clients then send that header instead of the login token, and every call fails with 401.
The login stores them encrypted in the authorization grant of your connection. The server uses them to call the Portkey API for you and to derive a private account id for usage metering. usefulapi does not show them on any page or in any reply. To stop all access, remove the connector and change or delete these credentials in Portkey.
Yes, if you approve it. 1 of the 16 Portkey tools can create or change data. The other 15 are marked read-only. Most MCP clients ask you to approve a tool call before it runs.
The Free plan gives 50 tool calls / month. Pro costs $9/mo or $90/yr, with 10,000 tool calls / month. Pro+ costs $39/mo or $390/yr, with 100,000 tool calls / month. Each subscription covers this Portkey server only. Cancel any time: your paid plan then runs to the end of the paid period. Run portkey_usage_status to see how many calls you used.
Ask the AI to run portkey_upgrade and choose Pro or Pro+, billed monthly or yearly: it returns a Stripe Checkout link. If you already have Pro, ask to change it to Pro+. After you confirm, your existing subscription changes in place, keeps its billing interval and charges the price difference for the rest of the period immediately. To cancel, run portkey_cancel_subscription. Your paid plan continues to the end of the paid period.
Tell the AI what you wanted to do. It can send the request with portkey_request_feature. We store the text with the server name, the type of AI client and the tool you tried, not with your account, and read every request when we plan new tools. You can send up to 5 requests per day.
Any client that supports remote MCP servers (Streamable HTTP) with OAuth login: Claude (web and desktop), Claude Code, ChatGPT, Codex, Cursor, VS Code, Windsurf and others.
This is a Model Context Protocol endpoint — meant to be connected from an AI client, not opened in a browser. An invalid_token response at the URL is the auth gate working as designed; clients authenticate automatically.
Need a tool this server does not have, or a connection to another system? We build custom integrations and MCP servers.