About Portkey MCP by usefulapi
Portkey MCP by usefulapi is a hosted server for Portkey users and AI gateway operators. It provides read access to analytics and administration data, plus a tool for logging request feedback.
Key features
- 01Read analytics graphs and groups
- 02Inspect configs and API keys
- 03Browse virtual keys and workspaces
- 04Review users, invites, and audit logs
- 05Create request feedback
Use cases
- 01Review gateway analytics from an MCP client
- 02Check workspace members and configuration
- 03Log feedback on a Portkey request
MCP support
Exposes Portkey analytics, configuration, account, and audit data as tools for MCP clients, along with a feedback-writing tool. Connect through the hosted endpoint and authenticate with Portkey credentials.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://portkey.usefulapi.io/mcpclaude mcp add --transport http io-usefulapi-portkey https://portkey.usefulapi.io/mcp{"mcpServers":{"io-usefulapi-portkey":{"type":"http","url":"https://portkey.usefulapi.io/mcp"}}}Open in Cursor{"servers":{"io-usefulapi-portkey":{"type":"http","url":"https://portkey.usefulapi.io/mcp"}}}Open in VS Code{"mcpServers":{"io-usefulapi-portkey":{"type":"http","url":"https://portkey.usefulapi.io/mcp"}}}{"mcpServers":{"io-usefulapi-portkey":{"serverUrl":"https://portkey.usefulapi.io/mcp"}}}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.
Tools
16| Parameter | Type | Required | Description |
|---|---|---|---|
| metric | enum | Yes | Which metric graph to return.cost | requests | tokens | latency | error-rate | errors | status-codes | requests-per-user | rescued-requests | cache-hit-rate | cache-hit-latency | users | weighted-feedback |
| configs | string | No | Comma-separated config slugs to scope the graph. |
| metadata | string | No | JSON string of metadata key/values to scope the graph. |
| total_units | string | No | Bucket size for the series, e.g. "day" or "hour". |
| virtual_keys | string | No | Comma-separated virtual key slugs to scope the graph. |
| time_of_generation_max | string | Yes | End of the window (ISO8601, format YYYY-MM-DDTHH:MM:SS±HH:MM). |
| time_of_generation_min | string | Yes | Start of the window (ISO8601, format YYYY-MM-DDTHH:MM:SS±HH:MM). |
Details
From the project READMEPortkey MCP by usefulapi
Manage Portkey AI-gateway admin and observability from Claude, Cursor, or any MCP client — read analytics, configs, virtual keys, workspaces, users and audit logs, and log request feedback. Hosted, no local install: connect with your Portkey API key.
Live endpoint: https://portkey.usefulapi.io/mcp · Homepage: https://usefulapi.io
Connect
- Claude (claude.ai, Desktop): open Customize → Connectors, click + Add → Add custom connector, and paste
https://portkey.usefulapi.io/mcp. - Claude Code:
claude mcp add --transport http portkey https://portkey.usefulapi.io/mcp, then run/mcpto log in. - VS Code: Add to VS Code.
- Cursor and other clients: add the URL as a remote MCP server:
{
"mcpServers": {
"portkey": {
"url": "https://portkey.usefulapi.io/mcp"
}
}
}
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, and the login page asks for your Portkey credentials.
Step-by-step setup, where to find your credentials, and FAQ: https://usefulapi.io/portkey/
On first connect you'll paste your Portkey API key (from the Portkey app → API Keys). It's validated, stored per-user, and scoped to you — no keys in config files.
Tools
| Tool | Type | What it does |
|---|---|---|
portkey_get_analytics_graph | read | Get analytics graph |
portkey_get_analytics_group | read | Get analytics group |
portkey_list_configs | read | List configs |
portkey_get_config | read | Get config |
portkey_list_api_keys | read | List API keys |
portkey_get_api_key | read | Get API key |
portkey_list_virtual_keys | read | List virtual keys |
portkey_get_virtual_key | read | Get virtual key |
portkey_list_workspaces | read | List workspaces |
portkey_get_workspace | read | Get workspace |
portkey_list_workspace_members | read | List workspace members |
portkey_list_users | read | List users |
portkey_get_user | read | Get user |
portkey_list_user_invites | read | List user invites |
portkey_list_audit_logs | read | List audit logs |
portkey_create_feedback | write | Create feedback |
portkey_usage_status | meta | Usage status (free-tier meter) |
portkey_request_feature | meta | Request a missing feature |
portkey_upgrade | meta | Upgrade to Pro (unlimited*) |
portkey_cancel_subscription | meta | Cancel the Pro subscription |
read tools are read-only; write tools mutate data (clients should confirm them); meta tools report usage, manage your subscription or send a feature request.
Pricing
| Plan | Price | Limit |
|---|---|---|
| Free | $0 | 50 tool calls / month |
| Pro (per user) | $9/mo or $90/yr (2 months free) | Unlimited* |
*Pro has a fair use limit of 10,000 tool calls a month on this server. For more, write to [email protected].
Pro covers this server only. Subscribe with portkey_upgrade (it returns a Stripe Checkout link). Cancel any time with portkey_cancel_subscription: Pro continues to the end of the paid period, with no refund for the current period, and running portkey_upgrade before then undoes the cancel. Or write to [email protected].
License
MIT — see LICENSE. Documentation only; the server is hosted.
Comments
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.