# China NMPA Medical Device Data (remote · mcp.agentdatacore.com) Structured China NMPA UDI data for AI agents: devices, registrants, models and registration numbers. - Trust score: 71/100 (medium) - Change this week: +4 - Registry status: active - Liveness: live - Owner verified: no - Last scored: 2026-09-27 ## Components - remote · `mcp.agentdatacore.com`: 71/100 (this document), [markdown](https://verifymcp.io/servers/huangjunsong404-crypto-china-nmpa-udi/mcp.md), [page](https://verifymcp.io/servers/huangjunsong404-crypto-china-nmpa-udi/mcp) ## Channel facts - Endpoint: `https://mcp.agentdatacore.com/mcp` - Transports: `streamable-http` - Auth: `none` - Version: `1.0.1` ## Trust breakdown How this component scores in each security and reliability category. Every signal is checked automatically against the live server, and we only credit what we can confirm. Scores are 0–100 per category. Scoring method: https://verifymcp.io/docs/scoring (what has changed: https://verifymcp.io/docs/scoring/changelog) Scored 2026-09-27. - **Endpoint Security**: 63/100 - The endpoint's TLS certificate is valid, in date, and uses a strong key. - No authorisation is required to call this server. Every tool declares its destructiveHint and none is destructive, so open access doesn't expose one. - HTTPS enforcement could not be verified: the plaintext port answered with HTTP 405, which proves neither a plaintext path nor enforcement. - HSTS check failed: the Strict-Transport-Security header is absent. - DNSSEC check failed: this domain isn't protected by DNSSEC. - **Transport & Reachability**: 100/100 - Verified streamable-http transport via a live MCP handshake. - **Schema Quality & AI Usability**: 66/100 - AI-judged instruction clarity (excellent). - Context-footprint check failed: tool/resource definitions use about 5757 tokens (~274/item across 21 items; 21 tools + 0 resources), over budget; trim descriptions and params. - Usage-examples check failed: none of the tools include examples. - **Stability & Change Management**: 40/100 - Stability observed for 12 of 30 days with no destabilising changes; credit accrues until the full window elapses. - **Tool Coverage**: 100/100 - 100% of tools have a non-trivial description (not blank, and not just the tool's name). - 100% of tool parameters carry a description. - **Tool Safety**: 100/100 - No prompt-injection markers were found in the server instructions, tool names or descriptions we captured. - We read all 21 captured tool definition(s), and no name or description among them implies an irreversible operation. - An AI judge read all 22 captured unit(s) of tool text and found none that tries to manipulate the model reading it. - **Capabilities**: 100/100 - Implements a supported MCP spec version (2025-11-25); the latest is 2026-07-28. ## Install ### How do I install the China NMPA Medical Device Data MCP server? China NMPA Medical Device Data is a hosted endpoint at https://mcp.agentdatacore.com/mcp, so there is nothing to install locally. Ready-made configuration for Claude, Cursor, VS Code, Codex and 5 more is on this page, copied from each client's own documentation. ### Claude ```bash claude mcp add --transport http huangjunsong404-crypto-china-nmpa-udi 'https://mcp.agentdatacore.com/mcp' ``` ### Cursor ```json { "mcpServers": { "huangjunsong404-crypto-china-nmpa-udi": { "url": "https://mcp.agentdatacore.com/mcp" } } } ``` ### VS Code ```json { "servers": { "huangjunsong404-crypto-china-nmpa-udi": { "type": "http", "url": "https://mcp.agentdatacore.com/mcp" } } } ``` ### Codex ```toml [mcp_servers.huangjunsong404-crypto-china-nmpa-udi] url = "https://mcp.agentdatacore.com/mcp" ``` ### opencode ```json { "$schema": "https://opencode.ai/config.json", "mcp": { "huangjunsong404-crypto-china-nmpa-udi": { "type": "remote", "url": "https://mcp.agentdatacore.com/mcp", "enabled": true } } } ``` ### OpenClaw ```bash openclaw mcp add huangjunsong404-crypto-china-nmpa-udi --url 'https://mcp.agentdatacore.com/mcp' --transport streamable-http ``` ### Hermes ```yaml mcp_servers: huangjunsong404-crypto-china-nmpa-udi: url: "https://mcp.agentdatacore.com/mcp" ``` ### Netclaw ```json { "McpServers": { "huangjunsong404-crypto-china-nmpa-udi": { "Transport": "http", "Url": "https://mcp.agentdatacore.com/mcp" } } } ``` ### Vellum ```bash assistant mcp add huangjunsong404-crypto-china-nmpa-udi -t streamable-http -u 'https://mcp.agentdatacore.com/mcp' ``` ### Other ```json { "mcpServers": { "huangjunsong404-crypto-china-nmpa-udi": { "type": "http", "url": "https://mcp.agentdatacore.com/mcp" } } } ``` The mcpServers block is a cross-client convention. Remote transports vary, so check your client's docs. ## Changelog Every change recorded for this component, newest first. Days that predate change tracking, or that we cannot explain, say so: "we were watching and nothing happened" and "we were not watching" are different claims. ### 2026-09-27 (score 71, +1) No change was recorded against any check on this day. Stability & Change Management went from 37 to 40. That category is still filling its 30-day observation window: 11 days of observed history at the previous scan, 12 at this one. The score rises as the window fills, whether or not the server changes. ### 2026-09-25 (score 70, +1) - [functional] We updated how we score, so this day's move reflects our rubric, not a change to the server ### 2026-09-23 (score 69, +1) No change was recorded against any check on this day. Stability & Change Management went from 23 to 27. That category is still filling its 30-day observation window: 7 days of observed history at the previous scan, 8 at this one. The score rises as the window fills, whether or not the server changes. ### 2026-09-21 (score 68, +1) No change was recorded against any check on this day. Stability & Change Management went from 17 to 20. That category is still filling its 30-day observation window: 5 days of observed history at the previous scan, 6 at this one. The score rises as the window fills, whether or not the server changes. ### 2026-09-19 (score 67, +1) No change was recorded against any check on this day. Stability & Change Management went from 10 to 13. That category is still filling its 30-day observation window: 3 days of observed history at the previous scan, 4 at this one. The score rises as the window fills, whether or not the server changes. ### 2026-09-17 (score 66, 0) - [security] Tool “get_mofcom_trade_stats” rewrote its description, which is the text the model reads - [security] Tool “get_customs_country_trade” rewrote its description, which is the text the model reads - [security] Tool “get_customs_commodity_trade” rewrote its description, which is the text the model reads - [cosmetic] “get_customs_country_trade” reworded the description of “period” - [cosmetic] “get_customs_country_trade” reworded the description of “flow” - [cosmetic] “get_customs_country_trade” reworded the description of “country” - [cosmetic] “get_customs_commodity_trade” reworded the description of “period” - [cosmetic] “get_customs_commodity_trade” reworded the description of “metric” - [cosmetic] “get_customs_commodity_trade” reworded the description of “flow” - [cosmetic] “get_customs_commodity_trade” reworded the description of “commodity” - [cosmetic] “get_mofcom_trade_stats” reworded the description of “metric” - [cosmetic] “get_mofcom_trade_stats” reworded the description of “period” - [cosmetic] “get_mofcom_trade_stats” reworded the description of “flow” - [cosmetic] “get_mofcom_trade_stats” reworded the description of “country” ### 2026-09-16 (score 66, +13) - [security improvement] Authorization: unverified → partial - [security] The server rewrote its instructions, which are the text every model session reads - [security] Tool “search_standard_metadata” rewrote its description, which is the text the model reads - [security] Tool “search_udi_category” rewrote its description, which is the text the model reads - [security] Tool “search_udi_products” rewrote its description, which is the text the model reads - [security] Tool “find_ccc_products” rewrote its description, which is the text the model reads - [security] Tool “find_ccc_standard_references” rewrote its description, which is the text the model reads - [security] Tool “find_country_institutions” rewrote its description, which is the text the model reads - [security] Tool “find_license_catalog_items” rewrote its description, which is the text the model reads - [security] Tool “get_ccc_rule_versions” rewrote its description, which is the text the model reads - [security] Tool “get_customs_commodity_trade” rewrote its description, which is the text the model reads - [security] Tool “get_customs_country_trade” rewrote its description, which is the text the model reads - [security] Tool “get_foreign_investment_stats” rewrote its description, which is the text the model reads - [security] Tool “get_gazette_items” rewrote its description, which is the text the model reads - [security] Tool “get_information_standard” rewrote its description, which is the text the model reads - [security] Tool “get_mofcom_trade_stats” rewrote its description, which is the text the model reads - [security] Tool “get_recall_record” rewrote its description, which is the text the model reads - [security] Tool “get_standard_metadata” rewrote its description, which is the text the model reads - [security] Tool “get_udi_company_products” rewrote its description, which is the text the model reads - [security] Tool “get_udi_device” rewrote its description, which is the text the model reads - [security] Tool “get_udi_registration_products” rewrote its description, which is the text the model reads - [security] Tool “search_policy_facts” rewrote its description, which is the text the model reads - [security] Tool “search_recalls” rewrote its description, which is the text the model reads - [functional regression] Schema quality: pass → fail - [functional improvement] Tool coverage: 0% → 100% - [functional improvement] Stability: unverified → 0.03 - [functional] Schema quality: fair → excellent - [functional] Server version: 1.0.0 → 1.0.1 - [cosmetic] “get_udi_company_products” reworded the description of “snapshot” - [cosmetic] “search_recalls” reworded the description of “snapshot” - [cosmetic] “search_standard_metadata” reworded the description of “limit” - [cosmetic] “search_standard_metadata” reworded the description of “offset” - [cosmetic] “search_standard_metadata” reworded the description of “query” - [cosmetic] “search_standard_metadata” reworded the description of “snapshot” - [cosmetic] “search_standard_metadata” reworded the description of “status” - [cosmetic] “search_udi_category” reworded the description of “category” - [cosmetic] “search_udi_category” reworded the description of “limit” - [cosmetic] “search_udi_category” reworded the description of “offset” - [cosmetic] “search_udi_category” reworded the description of “snapshot” - [cosmetic] “search_udi_products” reworded the description of “limit” - [cosmetic] “search_udi_products” reworded the description of “offset” - [cosmetic] “search_udi_products” reworded the description of “query” - [cosmetic] “search_udi_products” reworded the description of “snapshot” - [cosmetic] “find_ccc_products” reworded the description of “limit” - [cosmetic] “find_ccc_products” reworded the description of “offset” - [cosmetic] “find_ccc_products” reworded the description of “product” - [cosmetic] “find_ccc_products” reworded the description of “snapshot” - [cosmetic] “find_ccc_standard_references” reworded the description of “limit” - [cosmetic] “find_ccc_standard_references” reworded the description of “offset” - [cosmetic] “find_ccc_standard_references” reworded the description of “snapshot” - [cosmetic] “find_ccc_standard_references” reworded the description of “standard_number” - [cosmetic] “find_country_institutions” reworded the description of “country” - [cosmetic] “find_country_institutions” reworded the description of “institution” - [cosmetic] “find_country_institutions” reworded the description of “limit” - [cosmetic] “find_country_institutions” reworded the description of “offset” - [cosmetic] “find_country_institutions” reworded the description of “snapshot” - [cosmetic] “find_license_catalog_items” reworded the description of “limit” - [cosmetic] “find_license_catalog_items” reworded the description of “offset” - [cosmetic] “find_license_catalog_items” reworded the description of “product” - [cosmetic] “find_license_catalog_items” reworded the description of “snapshot” - [cosmetic] “get_ccc_rule_versions” reworded the description of “limit” - [cosmetic] “get_ccc_rule_versions” reworded the description of “offset” - [cosmetic] “get_ccc_rule_versions” reworded the description of “rule_number” - [cosmetic] “get_ccc_rule_versions” reworded the description of “snapshot” - [cosmetic] “get_customs_commodity_trade” reworded the description of “commodity” - [cosmetic] “get_customs_commodity_trade” reworded the description of “flow” - [cosmetic] “get_customs_commodity_trade” reworded the description of “limit” - [cosmetic] “get_customs_commodity_trade” reworded the description of “metric” - [cosmetic] “get_customs_commodity_trade” reworded the description of “offset” - [cosmetic] “get_customs_commodity_trade” reworded the description of “period” - [cosmetic] “get_customs_commodity_trade” reworded the description of “snapshot” - [cosmetic] “get_customs_country_trade” reworded the description of “country” - [cosmetic] “get_customs_country_trade” reworded the description of “flow” - [cosmetic] “get_customs_country_trade” reworded the description of “limit” - [cosmetic] “get_customs_country_trade” reworded the description of “offset” - [cosmetic] “get_customs_country_trade” reworded the description of “period” - [cosmetic] “get_customs_country_trade” reworded the description of “snapshot” - [cosmetic] “get_foreign_investment_stats” reworded the description of “country” - [cosmetic] “get_foreign_investment_stats” reworded the description of “limit” - [cosmetic] “get_foreign_investment_stats” reworded the description of “metric” - [cosmetic] “get_foreign_investment_stats” reworded the description of “offset” - [cosmetic] “get_foreign_investment_stats” reworded the description of “snapshot” - [cosmetic] “get_foreign_investment_stats” reworded the description of “year” - [cosmetic] “get_gazette_items” reworded the description of “document_number” - [cosmetic] “get_gazette_items” reworded the description of “issue” - [cosmetic] “get_gazette_items” reworded the description of “limit” - [cosmetic] “get_gazette_items” reworded the description of “offset” - [cosmetic] “get_gazette_items” reworded the description of “snapshot” - [cosmetic] “get_mofcom_trade_stats” reworded the description of “country” - [cosmetic] “get_mofcom_trade_stats” reworded the description of “flow” - [cosmetic] “get_mofcom_trade_stats” reworded the description of “limit” - [cosmetic] “get_mofcom_trade_stats” reworded the description of “metric” - [cosmetic] “get_mofcom_trade_stats” reworded the description of “offset” - [cosmetic] “get_mofcom_trade_stats” reworded the description of “period” - [cosmetic] “get_mofcom_trade_stats” reworded the description of “snapshot” - [cosmetic] “get_recall_record” reworded the description of “limit” - [cosmetic] “get_recall_record” reworded the description of “offset” - [cosmetic] “get_recall_record” reworded the description of “record_id” - [cosmetic] “get_recall_record” reworded the description of “snapshot” - [cosmetic] “get_standard_metadata” reworded the description of “limit” - [cosmetic] “get_standard_metadata” reworded the description of “offset” - [cosmetic] “get_standard_metadata” reworded the description of “snapshot” - [cosmetic] “get_standard_metadata” reworded the description of “standard_number” - [cosmetic] “get_udi_company_products” reworded the description of “company” - [cosmetic] “get_udi_company_products” reworded the description of “limit” - [cosmetic] “get_udi_company_products” reworded the description of “offset” - [cosmetic] “get_udi_device” reworded the description of “limit” - [cosmetic] “get_udi_device” reworded the description of “offset” - [cosmetic] “get_udi_device” reworded the description of “snapshot” - [cosmetic] “get_udi_device” reworded the description of “udi” - [cosmetic] “get_udi_registration_products” reworded the description of “limit” - [cosmetic] “get_udi_registration_products” reworded the description of “offset” - [cosmetic] “get_udi_registration_products” reworded the description of “registration_no” - [cosmetic] “get_udi_registration_products” reworded the description of “snapshot” - [cosmetic] “search_policy_facts” reworded the description of “limit” - [cosmetic] “search_policy_facts” reworded the description of “offset” - [cosmetic] “search_policy_facts” reworded the description of “query” - [cosmetic] “search_policy_facts” reworded the description of “snapshot” - [cosmetic] “search_recalls” reworded the description of “limit” - [cosmetic] “search_recalls” reworded the description of “offset” - [cosmetic] “search_recalls” reworded the description of “query” ### 2026-09-15 (score 53) First indexed and scored. ## MCP tools (21) ### `get_udi_device` (~237 tokens) Look up medical-device records when a UDI identifier is known. Inputs: udi. Returns paginated structured records with source_class, provenance, snapshot, pagination and limits. Recorded facts only; this query performs no new inference or aggregation. Coverage is limited to the available snapshot. No match returns an empty data array. Required: udi. Missing required or invalid input produces an argument error. Input parameters: - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… - `udi` (string, required): Recorded UDI identifier; retain leading zeros. Exact match. Non-empty string, at most 256 characters; control characters are not accepted. Required. ### `search_udi_products` (~237 tokens) Find medical devices by product or trade-name text. Inputs: query. Returns paginated structured records with source_class, provenance, snapshot, pagination and limits. Recorded facts only; this query performs no new inference or aggregation. Coverage is limited to the available snapshot. No match returns an empty data array. Required: query. Missing required or invalid input produces an argument error. Input parameters: - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `query` (string, required): Recorded product or trade-name text. Literal, case-sensitive substring match; no fuzzy search. Non-empty string, at most 256 characters; control characters are not accepted. Required. - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… ### `get_udi_company_products` (~246 tokens) Find recorded medical-device products associated with a company or registrant name. Inputs: company. Returns paginated structured records with source_class, provenance, snapshot, pagination and limits. Recorded facts only; this query performs no new inference or aggregation. Coverage is limited to the available snapshot. No match returns an empty data array. Required: company. Missing required or invalid input produces an argument error. Input parameters: - `company` (string, required): Recorded registrant/company name, including recorded English names. Literal, case-sensitive substring match; no fuzzy search. Non-empty string, at most 256 characters; control characters are not acce… - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… ### `get_udi_registration_products` (~239 tokens) Find medical-device records associated with a registration or filing number. Inputs: registration_no. Returns paginated structured records with source_class, provenance, snapshot, pagination and limits. Recorded facts only; this query performs no new inference or aggregation. Coverage is limited to the available snapshot. No match returns an empty data array. Required: registration_no. Missing required or invalid input produces an argument error. Input parameters: - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `registration_no` (string, required): Recorded registration or filing number, including its original punctuation. Exact match. Non-empty string, at most 256 characters; control characters are not accepted. Required. - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… ### `search_udi_category` (~240 tokens) Find medical devices by a recorded category name or classification code. Inputs: category. Returns paginated structured records with source_class, provenance, snapshot, pagination and limits. Recorded facts only; this query performs no new inference or aggregation. Coverage is limited to the available snapshot. No match returns an empty data array. Required: category. Missing required or invalid input produces an argument error. Input parameters: - `category` (string, required): Recorded device category name or classification code. Literal, case-sensitive substring match; no fuzzy search. Non-empty string, at most 256 characters; control characters are not accepted. Required. - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… ### `search_recalls` (~236 tokens) Find recall notices and detail records by product, manufacturer or title text. Inputs: query. Returns paginated structured records with source_class, provenance, snapshot, pagination and limits. Recorded notices only; an empty result does not establish absence of recalls. No match returns an empty data array. Required: query. Missing required or invalid input produces an argument error. Input parameters: - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `query` (string, required): Recorded recall title, product name or manufacturer text. Literal, case-sensitive substring match; no fuzzy search. Non-empty string, at most 256 characters; control characters are not accepted. Requ… - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… ### `get_recall_record` (~233 tokens) Retrieve a recall notice or detail record using an existing search-result identifier. Inputs: record_id. Returns paginated structured records with source_class, provenance, snapshot, pagination and limits. Recorded facts only; this query performs no new inference or aggregation. Coverage is limited to the available snapshot. No match returns an empty data array. Required: record_id. Missing required or invalid input produces an argument error. Input parameters: - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `record_id` (integer, required): Existing integer recall record identifier returned by search_recalls. Exact match. Non-negative integer within the server integer range. Required. - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… ### `find_ccc_products` (~229 tokens) Find CCC catalog product entries by product or category text. Inputs: product. Returns paginated structured records with source_class, provenance, snapshot, pagination and limits. Catalog matches only; does not determine certification or legal applicability. No match returns an empty data array. Required: product. Missing required or invalid input produces an argument error. Input parameters: - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `product` (string, required): Recorded CCC product or category text. Literal, case-sensitive substring match; no fuzzy search. Non-empty string, at most 256 characters; control characters are not accepted. Required. - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… ### `get_ccc_rule_versions` (~232 tokens) Find recorded versions and dates for a CCC rule number. Inputs: rule_number. Returns paginated structured records with source_class, provenance, snapshot, pagination and limits. Recorded version facts only; does not infer legal applicability. No match returns an empty data array. Required: rule_number. Missing required or invalid input produces an argument error. Input parameters: - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `rule_number` (string, required): CCC rule number exactly as recorded, including punctuation and version suffix if present. Exact match. Non-empty string, at most 256 characters; control characters are not accepted. Required. - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… ### `find_ccc_standard_references` (~236 tokens) Find occurrences of a recorded standard number in CCC reference metadata. Inputs: standard_number. Returns paginated structured records with source_class, provenance, snapshot, pagination and limits. Reference occurrences only; does not imply equivalence or applicability. No match returns an empty data array. Required: standard_number. Missing required or invalid input produces an argument error. Input parameters: - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… - `standard_number` (string, required): Standard number exactly as recorded, including spaces, punctuation and version suffix if present. Exact match. Non-empty string, at most 256 characters; control characters are not accepted. Required. ### `get_standard_metadata` (~228 tokens) Retrieve standard metadata when its recorded standard number is known. Inputs: standard_number. Returns paginated structured records with source_class, provenance, snapshot, pagination and limits. Metadata only; no standard full text. No match returns an empty data array. Required: standard_number. Missing required or invalid input produces an argument error. Input parameters: - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… - `standard_number` (string, required): Standard number exactly as recorded, including spaces, punctuation and version suffix if present. Exact match. Non-empty string, at most 256 characters; control characters are not accepted. Required. ### `search_standard_metadata` (~294 tokens) Find standard metadata by title text, optionally filtered by its recorded status. Inputs: query, status. Returns paginated structured records with source_class, provenance, snapshot, pagination and limits. Metadata only; no standard full text. No match returns an empty data array. Required: query. Missing required or invalid input produces an argument error. Input parameters: - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `query` (string, required): Recorded Chinese or English standard-title text. Literal, case-sensitive substring match; no fuzzy search. Non-empty string, at most 256 characters; control characters are not accepted. Required. Sup… - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… - `status` (string): Standard status label exactly as published; no translation or status inference. Exact match. Non-empty string, at most 256 characters; control characters are not accepted. Optional; omitted means no… ### `get_foreign_investment_stats` (~365 tokens) Find foreign-investment observations for a statistical year, optionally filtered by country or measure. Inputs: year, country, metric. Returns paginated structured records with source_class, provenance, snapshot, pagination and limits. Recorded facts only; this query performs no new inference or aggregation. Coverage is limited to the available snapshot. No match returns an empty data array. Required: year. Missing required or invalid input produces an argument error. Input parameters: - `country` (string): Country or region label exactly as recorded; no country-code or alias conversion. Exact match. Non-empty string, at most 256 characters; control characters are not accepted. Optional; omitted means n… - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `metric` (string): Statistical measure label exactly as recorded; no unit conversion or aggregation. Exact match. Non-empty string, at most 256 characters; control characters are not accepted. Optional; omitted means n… - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… - `year` (integer, required): Recorded statistical year as an integer; this is not the publication year. Exact match. Non-negative integer within the server integer range. Required. Supplied filters are combined with AND. ### `find_country_institutions` (~312 tokens) Find recorded institutions and roles for a country, optionally filtered by institution text. Inputs: country, institution. Returns paginated structured records with source_class, provenance, snapshot, pagination and limits. Recorded facts only; this query performs no new inference or aggregation. Coverage is limited to the available snapshot. No match returns an empty data array. Required: country. Missing required or invalid input produces an argument error. Input parameters: - `country` (string, required): Country or region label exactly as recorded; no country-code or alias conversion. Exact match. Non-empty string, at most 256 characters; control characters are not accepted. Required. Supplied filter… - `institution` (string): Recorded institution name or entity identifier text. Literal, case-sensitive substring match; no fuzzy search. Non-empty string, at most 256 characters; control characters are not accepted. Optional;… - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… ### `get_gazette_items` (~295 tokens) Find entries in a gazette issue, optionally filtered by document number. Inputs: issue, document_number. Returns paginated structured records with source_class, provenance, snapshot, pagination and limits. Recorded facts only; this query performs no new inference or aggregation. Coverage is limited to the available snapshot. No match returns an empty data array. Required: issue. Missing required or invalid input produces an argument error. Input parameters: - `document_number` (string): Document number exactly as recorded. Exact match. Non-empty string, at most 256 characters; control characters are not accepted. Optional; omitted means no filter on this field. Supplied filters are… - `issue` (string, required): Gazette issue label exactly as recorded. Exact match. Non-empty string, at most 256 characters; control characters are not accepted. Required. Supplied filters are combined with AND. - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… ### `search_policy_facts` (~240 tokens) Find policy and document facts by title or document-number text. Inputs: query. Returns paginated structured records with source_class, provenance, snapshot, pagination and limits. Recorded facts only; this query performs no new inference or aggregation. Coverage is limited to the available snapshot. No match returns an empty data array. Required: query. Missing required or invalid input produces an argument error. Input parameters: - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `query` (string, required): Recorded policy title or document-number text. Literal, case-sensitive substring match; no fuzzy search. Non-empty string, at most 256 characters; control characters are not accepted. Required. - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… ### `find_license_catalog_items` (~238 tokens) Find licence catalog entries by recorded product or commodity text. Inputs: product. Returns paginated structured records with source_class, provenance, snapshot, pagination and limits. Recorded facts only; this query performs no new inference or aggregation. Coverage is limited to the available snapshot. No match returns an empty data array. Required: product. Missing required or invalid input produces an argument error. Input parameters: - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `product` (string, required): Recorded licence commodity or subject text. Literal, case-sensitive substring match; no fuzzy search. Non-empty string, at most 256 characters; control characters are not accepted. Required. - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… ### `get_mofcom_trade_stats` (~414 tokens) Query collected country-workbook trade observations (flows relative to the named reporting country) or China–Portuguese-speaking-country bilateral year-to-date observations. Inputs: period and optional country, flow and metric. This collection is not China national all-partner monthly totals. Returns original structured observations, units, scope, source_class, provenance, snapshot, pagination and limits. No conversion, aggregation or cross-period calculation. No match returns an empty data array. Required: period. Missing required or invalid input produces an argument error. Input parameters: - `country` (string): Reporting country in country workbooks, or China bilateral partner in Portuguese-speaking-country observations; returned scope distinguishes them. Exact recorded name or a documented one-to-one alias… - `flow` (string): Flow relative to the named reporting country for COUNTRY_WORKBOOK_AS_REPORTED; China flow relative to the named partner for CHINA_PORTUGUESE_COUNTRIES_YTD. Recorded values IMPORT, EXPORT or BOTH; sco… - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `metric` (string): Recorded metric trade_value; country workbook unit is 亿美元, China–Portuguese-speaking-partner unit is 万美元. No unit conversion. Optional. - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `period` (string, required): Recorded statistical period: YYYY-MM for a single month, or YYYY-01/YYYY-MM for January-to-month cumulative observations. Exact collected period; these periods are not interchangeable. A snapshot ide… - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… ### `get_customs_country_trade` (~353 tokens) Query China merchandise trade by partner country/region, including all-partner totals using country TOTAL. Inputs: period, country and optional flow. IMPORT and EXPORT are China imports and exports; BOTH is recorded combined trade. Returns recorded trade values and units (currently 1000USD), scope, source_class, provenance, snapshot, pagination and limits. No commodity filter or currency conversion. No match returns an empty data array. Required: period, country. Missing required or invalid input produces an argument error. Input parameters: - `country` (string, required): China trade partner country/region, or TOTAL for all partners. TOTAL is an aggregate, not an alias for China. Exact recorded name or selected one-to-one alias; original labels remain in output. Requi… - `flow` (string): China trade direction: IMPORT or EXPORT or BOTH (recorded combined total). Optional; omission includes collected directions. - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `period` (string, required): Recorded statistical period: YYYY-MM for a single month, or YYYY-01/YYYY-MM for January-to-month cumulative observations. Exact collected period; these periods are not interchangeable. A snapshot ide… - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… ### `get_customs_commodity_trade` (~404 tokens) Query China import or export observations by recorded major-commodity label. Inputs: period, commodity, flow and optional metric. Selected exact Chinese aliases and normalized existing names are supported; this is not general translation or HS-code classification. Returns trade_value or quantity observations with original commodity labels, units, scope, source_class, provenance, snapshot, pagination and limits. No category merging or conversion. No match returns an empty data array. Required: period, commodity, flow. Missing required or invalid input produces an argument error. Input parameters: - `commodity` (string, required): Recorded major-commodity name or a selected exact Chinese alias. Existing literal substring matching remains; known aliases resolve only to their exact canonical label. NFKC, leading/trailing spaces… - `flow` (string, required): China trade direction: IMPORT or EXPORT. Required. - `limit` (integer): Maximum number of records returned. Integer 1–100; default 20. A page may contain fewer records. - `metric` (string): Recorded metric trade_value or quantity. trade_value currently uses 1000USD; quantity units vary by commodity and are returned unchanged. Optional; omission includes both collected metrics. - `offset` (integer): Zero-based pagination offset. Integer 0–999; default 0. Offsets of 1000 or greater are rejected. Use the returned next_offset when present. - `period` (string, required): Recorded statistical period: YYYY-MM for a single month, or YYYY-01/YYYY-MM for January-to-month cumulative observations. Exact collected period; these periods are not interchangeable. A snapshot ide… - `snapshot` (string): Optional expected snapshot.version, not a historical-version selector. If omitted, the current dataset snapshot is used. For continued pagination reuse the previous response snapshot.version; a misma… ### `get_information_standard` (~42 tokens) Retrieve definitions of information formation classes, provenance fields and snapshot/version metadata. Takes no arguments. Returns the information-standard document, not business records or an assessment of facts. ## Diagnostics Captured diagnostic sections: TLS, DNSSEC, Authorisation, Transports. The full working is on the page: https://verifymcp.io/servers/huangjunsong404-crypto-china-nmpa-udi/mcp#diagnostics ## Score history - 2026-09-27: 71 - 2026-09-26: 70 - 2026-09-25: 70 - 2026-09-24: 69 - 2026-09-23: 69 - 2026-09-22: 68 - 2026-09-21: 68 - 2026-09-20: 67 - 2026-09-19: 67 - 2026-09-18: 66 - 2026-09-17: 66 - 2026-09-16: 66 - 2026-09-15: 53 ## Common questions ### What is the China NMPA Medical Device Data MCP server? China NMPA Medical Device Data is an MCP server listed in the public MCP registry as io.github.huangjunsong404-crypto/china-nmpa-udi. Structured China NMPA UDI data for AI agents: devices, registrants, models and registration numbers. This page covers its hosted endpoint (https://mcp.agentdatacore.com/mcp). ### Is the China NMPA Medical Device Data MCP server safe to use? China NMPA Medical Device Data scores 71 out of 100 on VerifyMCP. That is a record of what we were able to check automatically, not an endorsement. The category breakdown on this page shows every signal behind the number, including the ones we could not confirm. ### What tools does the China NMPA Medical Device Data MCP server expose? China NMPA Medical Device Data exposes 21 tools: get_udi_device, search_udi_products, get_udi_company_products, get_udi_registration_products, search_udi_category, and 16 more. Their descriptions and schemas cost roughly 5,550 tokens of context every time the server is loaded. ### Does the China NMPA Medical Device Data MCP server require authentication? No. We connected to China NMPA Medical Device Data without credentials and it answered, so anything it exposes is reachable by anyone who knows the address. ### Is the China NMPA Medical Device Data MCP server still maintained? China NMPA Medical Device Data is still listed as active in the MCP registry. We last reached this channel on 27 September 2026. Those dates come from our own scans of the registry and the channel itself, not from anything the publisher announced. ## Links - Remote endpoint: https://mcp.agentdatacore.com/mcp - Repository: https://github.com/huangjunsong404-crypto/agentdata-api - Changelog RSS feed: https://verifymcp.io/servers/huangjunsong404-crypto-china-nmpa-udi/mcp.xml - Changelog JSON feed: https://verifymcp.io/servers/huangjunsong404-crypto-china-nmpa-udi/mcp.json - HTML version of this page: https://verifymcp.io/servers/huangjunsong404-crypto-china-nmpa-udi/mcp