Patents & Trademarks MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| USPTO_IP_RPS | No | Per-IP token bucket rate on the HTTP transport. | 5 |
| USPTO_API_KEY | No | ODP API key, sent as X-API-KEY. Unset means the odp source is disabled. | |
| USPTO_SANDBOX | No | Sandbox runner to use: auto, isolate, or deno. | auto |
| USPTO_IP_BURST | No | Per-IP token bucket burst on the HTTP transport. | 20 |
| USPTO_USER_AGENT | No | User agent used to identify your deployment. | uspto-mcp-server/0.1 … |
| USPTO_AUTH_ISSUER | No | OIDC issuer. Required for OIDC providers. | |
| USPTO_AUTH_SCOPES | No | Comma-separated scopes every token must carry; advertised in resource metadata. | |
| USPTO_CACHE_TTL_MS | No | Response cache TTL in milliseconds. | 600000 |
| USPTO_TSDR_API_KEY | No | TSDR API key, sent as USPTO-API-KEY. Unset means the tsdr source is disabled. | |
| USPTO_ALLOWED_HOSTS | No | Extra accepted Host values, comma-separated. A value of * disables the check (unsafe). | |
| USPTO_AUTH_AUDIENCE | No | OIDC audience. Required for OIDC providers. | |
| USPTO_AUTH_JWKS_URL | No | OIDC JWKS URL over https (http only on loopback). Required for OIDC providers. | |
| USPTO_AUTH_PROVIDER | No | Authentication provider: none (only with --insecure), generic-oidc, clerk, workos, auth0, or accounts. Unknown values stop startup. | none |
| USPTO_PUBLIC_ORIGIN | No | Public origin clients reach, used for OAuth metadata and trusted Host. Required for non-loopback HTTP unless USPTO_ALLOWED_HOSTS is set. | |
| USPTO_ALLOWED_ORIGINS | No | Allowed browser Origin values, comma-separated, in addition to the server's own and loopback origins. | |
| USPTO_CACHE_MAX_ITEMS | No | Maximum number of items in the response cache. | 1000 |
| USPTO_ODP_CONCURRENCY | No | Number of ODP requests in flight per key. Keep at 1. | 1 |
| USPTO_CLIENT_IP_HEADER | No | Header your proxy overwrites with the client address (for Railway: x-real-ip). | |
| USPTO_ODP_RATE_PER_SEC | No | Number of ODP requests started per second per key. | 4 |
| USPTO_TSDR_CONCURRENCY | No | Number of TSDR requests in flight per key. Keep at 1. | 1 |
| USPTO_ODP_RATE_PER_HOUR | No | Hourly ODP attempts per configured key (per account in account mode). Must be at least 1. | 5000 |
| USPTO_TSDR_RATE_PER_SEC | No | Number of TSDR requests started per second per key. | 1 |
| USPTO_MAX_RESPONSE_BYTES | No | Largest upstream response body accepted. | 8388608 |
| USPTO_TSDR_RATE_PER_HOUR | No | Hourly TSDR attempts per configured key (per account in account mode). Must be at least 1. | 3000 |
| USPTO_SUBJECT_DAILY_QUOTA | No | Requests per authenticated subject per UTC day. | 10000 |
| USPTO_MAX_CONCURRENT_EXECUTIONS | No | Maximum simultaneous executions per server. Beyond this returns Busy. | 8 |
| USPTO_ODP_MAX_CALLS_PER_EXECUTE | No | Maximum upstream ODP calls per execute() call. Exceeding it returns CallBudgetExceeded. | 50 |
| USPTO_ODP_SUBJECT_RATE_PER_HOUR | No | Per-authenticated-subject ODP requests per hour. HTTP only. | 1000 |
| USPTO_MAX_EXECUTIONS_PER_SUBJECT | No | Maximum simultaneous executions per authenticated subject. Beyond this returns Busy. | 2 |
| USPTO_TSDR_MAX_CALLS_PER_EXECUTE | No | Maximum upstream TSDR calls per execute() call. Exceeding it returns CallBudgetExceeded. | 25 |
| USPTO_TSDR_SUBJECT_RATE_PER_HOUR | No | Per-authenticated-subject TSDR requests per hour. HTTP only. | 600 |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
USPTO data through code mode. Workflow: search_api (find the SDK method and field names) -> describe_schema (exact signature) -> execute (run plain JavaScript against the odp and tsdr globals).
odp = USPTO Open Data Portal: odp.patents.* (patent applications/file wrapper), odp.ptab.* (IPR/PGR trials, appeals, interferences), odp.petitions., odp.officeActions., odp.oa.* (legacy office action datasets), odp.bulk.* (bulk dataset catalog). tsdr = trademark status by serial or registration number.
USPTO allows one request at a time per API key and has weekly quotas: use one well-filtered search with fields and a sensible limit rather than many lookups, and return only the values you need.
Application numbers (e.g. 17/248,024) differ from patent numbers (e.g. 11,646,472): resolve a patent number with odp.patents.byPatentNumber first.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| search_apiA | BM25 search over the SDK methods and response-field dictionary for all bound sources (odp, tsdr): patent applications (file wrapper), PTAB trials/appeals/interferences, petition decisions, office actions, bulk datasets, and trademark status. Returns method signatures and JavaScript examples for use in execute. Start here. |
| describe_schemaA | Look up an endpoint or field by exact dotted id (path) or by namespace prefix. Use to drill into the bound source surfaces (odp., tsdr.). |
| executeA | Run JavaScript inside a sandbox (no net, fs, env, or subprocess). Plain JS only — TypeScript type annotations are rejected. Globals: odp, tsdr. Top-level await is allowed; |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 3 tools
The three tools map to distinct phases: discovery by fuzzy search (search_api), exact/prefix schema lookup (describe_schema), and execution (execute). The overlap between search_api and describe_schema is real — both expose the same API surface — but the descriptions give clear routing cues ('Start here' vs 'drill into ... by exact dotted id'), so misselection is unlikely.
Two tools follow a verb_noun pattern (search_api, describe_schema) and the third is a bare verb (execute), which is a minor but defensible deviation since it is the terminal action. All names are lowercase snake_case and readable; there is no mixing of camelCase or arbitrary styles.
Three tools is at the low end of the ideal range but fits a code-execution gateway, where an entire SDK is reached through `execute` rather than one tool per endpoint. Splitting discovery into search and describe is justified by the breadth of the odp/tsdr surface; nothing feels padded.
The lifecycle of discover -> inspect -> call is fully covered, and arbitrary JS execution means no API operation is a dead end. Minor gaps exist: no dry-run or validation step before `execute`, and no way to inspect rate-limit/queue state despite the documented one-request-per-key constraint.