mcp-yuntrack
Track parcels via YunTrack using a single tracking number or batch lookups.
Track a single parcel: Provide one
trackingIdto retrieve its current status and details.Track multiple parcels: Provide an array of
trackingIds(up to 3 per call) to retrieve tracking information for all of them in parallel.Raw API data: Returns the raw JSON response directly from the YunTrack Query API (
services.yuntrack.com/Track/Query) without transformation.Fast & efficient: Uses a persistent headless Chromium browser to bypass WAF/TLS fingerprinting, with response times around ~400 ms. The browser context is reused across calls for efficiency.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@mcp-yuntracktrack package UJ123456789SE"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
mcp-parceltrack
MCP server (stdio transport) for parcel tracking, supporting YunTrack (YunExpress) and 4PX. You do not pick a carrier — the server works it out from the tracking number.
How it works
Each carrier is a provider behind a common interface, and a router chooses per
tracking number: a recognised prefix (4PX…, or YT/UJ/BCM/0099) goes straight to
its carrier, while an unrecognised one is probed cheapest-carrier-first.
The two carriers could hardly be less alike. 4PX is a plain open endpoint (~200 ms,
no browser, no signature) — but it only ever reads the first entry of its
queryCodes array, so it cannot batch: each parcel is its own request, fanned out
a few at a time.
YunTrack is the opposite. It batches 100 numbers per call, but its provider has to
call services.yuntrack.com/Track/Query with a signed POST request (HMAC-SHA256,
key found in the page bundle). Because Alibaba Cloud WAF blocks
requests that lack Chrome's TLS fingerprint, a headless Chromium browser is kept alive
in the background and all API calls are made via page.evaluate() — the browser's
own fetch() — rather than loading the full tracking page each time. This keeps
response times around 400 ms per query.
A shared browser context is reused across all calls so warm-up only happens once, and it is started lazily — a 4PX-only lookup never launches Chromium at all. The raw JSON from each API is returned as-is — no field interpretation.
Related MCP server: Parcel MCP Server
Build
npm install
npx playwright install chromium
npm run buildDevelopment & testing
npm run inspect # opens MCP Inspector in the browser — lets you call tools interactively
npm run dev # watch mode (tsc + node --watch)Installing in Claude Desktop
Open (or create)
~/Library/Application Support/Claude/claude_desktop_config.jsonAdd the server under
mcpServers:
{
"mcpServers": {
"parceltrack": {
"command": "/absolute/path/to/mcp-parceltrack/node_modules/.bin/tsx",
"args": ["/absolute/path/to/mcp-parceltrack/src/index.ts"]
}
}
}Restart Claude Desktop — the
track_parceltool is now available.
Installing in Claude Code (CLI)
Run once from any directory:
claude mcp add parceltrack /absolute/path/to/mcp-parceltrack/node_modules/.bin/tsx /absolute/path/to/mcp-parceltrack/src/index.tsRestart Claude Code. The tool is available in every project.
To confirm it loaded:
claude mcp listInstalling in Claude Code via project config
Add a .claude/mcp.json (or mcp.json) in your project root:
{
"mcpServers": {
"parceltrack": {
"command": "/absolute/path/to/mcp-parceltrack/node_modules/.bin/tsx",
"args": ["/absolute/path/to/mcp-parceltrack/src/index.ts"]
}
}
}This makes the server available to everyone who opens the project in Claude Code.
Tool: track_parcel
Single tracking number:
{ "trackingId": "UJ123456789SE" }Batch — carriers may be mixed freely:
{ "trackingIds": ["UJ123456789SE", "4PX3003133457168CN"] }Forcing a carrier (rarely needed; only when auto-detection is known to be wrong):
{ "trackingId": "1234567890", "provider": "4px" }Result
One entry per tracking number, in the order you gave them:
[
{
"trackingId": "4PX3003133457168CN",
"provider": "4px",
"found": true,
"result": { "queryCode": "4PX3003133457168CN", "tracks": [ ... ] }
}
]provider— which carrier answered.found— whether that carrier actually has a record of the parcel. Check this, notresult: both carriers reply to numbers they have never heard of (4PX with an empty stub, YunTrack by echoing the number back with zeroed fields), so a payload alone proves nothing.result— the carrier's raw JSON, unmodified. Shapes differ per carrier.error— present only when the request itself failed, never for a simple miss.
Checking the routing
npm run check:routingHits both carriers live and asserts the routing rules (correct carrier per prefix, input order preserved, no browser launched for a 4PX-only lookup).
The structural checks run on throwaway numbers. To also prove that a real parcel
comes back as found, pass tracking numbers of your own — none are committed:
PARCELTRACK_TEST_4PX=4PX…CN PARCELTRACK_TEST_YUNTRACK=YT… npm run check:routingAvailable Tools
1 tooltrack_parcelA
Track one or more parcels via YunTrack. Returns the raw JSON from the YunTrack Query API.
| Name | Required | Description | Default |
|---|---|---|---|
| trackingId | No | A single tracking number to look up. | |
| trackingIds | No | Multiple tracking numbers to look up in parallel (max 3 at a time). |
TDQS
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, placing full burden on the description. It discloses the return format (raw JSON) and implicitly indicates a read operation, but does not mention authentication requirements, rate limits, or error handling. Adequate but incomplete.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two concise sentences with no wasted words. The first states the action, the second describes the output format. Excellent front-loading of key information.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's simplicity (2 params, no output schema, no siblings), the description provides the essential purpose and return type. However, it lacks guidance on edge cases (invalid tracking IDs) and assumes familiarity with YunTrack API.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100%, so the baseline is 3. The description adds 'via YunTrack' context but does not enhance understanding of the parameters beyond what the schema already provides (single vs multiple tracking numbers, max 3).
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool tracks parcels via YunTrack, with explicit reference to single or multiple tracking numbers and the return format (raw JSON). This is specific and unambiguous.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No alternative tools are provided, so explicit when-to-use guidance is less critical. However, the description does not explain when to use this tool specifically, e.g., for checking parcel status versus other tracking methods. Implies usage but lacks context.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections. Dates show when Glama detected each change.
1 tool update
v1.0.0- First observed
track_parcel
TDQS
With only one tool, there is no ambiguity between tools. The tool's purpose is clearly defined as tracking parcels via YunTrack.
The single tool 'track_parcel' follows a clear verb_noun pattern using snake_case, which is consistent within itself.
One tool is quite thin for a typical MCP server, but it may be appropriate if the server is narrowly scoped to parcel tracking only. The tool handles multiple parcels in one call, reducing the need for additional tools.
For a simple parcel tracking server, the single tool covers the core functionality. Minor gaps might exist, such as listing carriers or getting tracking history, but these are not essential for basic tracking.
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Connectors
Track packages across 1,300+ global carriers with real-time status and AI-powered delivery dates.
Track parcels across 2,500+ carriers and 3PLs: live delivery status, event history, carrier lookup.
Real-time order tracking for Shopify merchants: look up, list, and refresh shipments.
Track shipments and search shipping products (DPD, InPost, DHL, FedEx, UPS, GLS) via GlobKurier API.
Related MCP Servers
- AlicenseBqualityDmaintenanceAllows tracking of parcel deliveries across various carriers using the 17track.net API, with automatic or manual carrier detection.2216MIT
- FlicenseNot gradedqualityBmaintenanceEnables AI assistants to track and manage deliveries by interacting with the Parcel delivery tracking API. Users can add new shipments, retrieve active delivery statuses, and look up carrier information through natural language.2-
- AlicenseNot gradedqualityBmaintenanceEnables AI to query your personal package tracking information from Taobao, JD, and Cainiao after a one-time QR code login.1MIT
- FlicenseNot gradedqualityDmaintenanceEnables package tracking across multiple carriers by auto-detecting the carrier from a tracking number and returning structured events. Supports USPS, UPS, FedEx, DHL, India Post, Delhivery, BlueDart, and Aramex.-
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/waldell/mcp-parceltrack'
If you have feedback or need assistance with the MCP directory API, please join our Discord server