Shipping Cost MCP Connector
Pulls real and estimated shipping costs from Shiprocket and Delhivery into Claude, with rollup summaries for yesterday, last 7 days, MTD, and last 30 days.
README
Shipping Cost MCP Connector
Pulls Shiprocket (actual) + Delhivery (estimated proxy) shipping costs into Claude, with Yesterday / Last 7 Days / MTD / Last 30 Days rollups.
Important limitations (read first)
- Shiprocket: costs come from live order data via the documented
/ordersAPI. The exact field name for shipping charge can vary by account/plan. Before trusting numbers, run theshiprocket_raw_order_sampletool and check which field (shipping_charges,freight_charges,charges, etc.) actually holds the value on your account, then adjustCHARGE_FIELD_CANDIDATESinlib/shiprocket.jsif needed. - Delhivery: there's no public API that lists your historical shipments by date,
and no API that returns actual billed amounts — only a live rate-estimate endpoint.
So
delhivery_cost_estimaterequires you to supply the shipment list (date, origin/dest pincode, weight) — e.g. exported from your own order records — and it estimates what each shipment should cost. It will be close but not exact to the paisa, same as your own Delhivery invoice reconciliation would show.
Setup
-
Generate Shiprocket API credentials Shiprocket panel → Settings → API → Add New API User → note the email + password emailed to that address.
-
Generate Delhivery API token Delhivery One / CL Panel → Settings → API Setup → copy your static token.
-
Deploy to Vercel
vercel deploy -
Set environment variables in the Vercel project dashboard (not in code):
SHIPROCKET_API_EMAILSHIPROCKET_API_PASSWORDDELHIVERY_API_TOKENDELHIVERY_CLIENT_NAME(your Delhivery client/company name as registered)MCP_ACCESS_TOKEN(any random string you choose — this now doubles as both the query-param token for manual testing AND the OAuth access token issued to Claude, plus the HMAC signing secret for authorization codes)
-
Add the connector in Claude using the base URL — you no longer need to append
?token=...for Claude itself (it now goes through the OAuth flow below automatically), though the query-param path still works for curl/manual testing:https://<your-vercel-app>.vercel.app/api/mcp
Why there's an OAuth layer here
Claude's custom connector UI only supports OAuth (it has no field for a plain
Bearer/query token), and its connection flow always attempts Dynamic Client
Registration (RFC 7591) against your server before calling any tools. A plain
token-only MCP server has no /register endpoint, so that step 404s and Claude
shows "couldn't register with sign-in service." To work around this, this
project includes a minimal, stateless OAuth 2.1 shim:
/.well-known/oauth-authorization-server— discovery metadata/.well-known/oauth-protected-resource— resource-first discovery metadata/api/register— auto-issues a client_id, no persistence needed/api/authorize— auto-approves and redirects back immediately (no login screen, since this server only ever serves you)/api/token— verifies the authorization code (HMAC-signed, PKCE-checked) and hands back your existingMCP_ACCESS_TOKENas the access token
Nothing here is persisted server-side; the "codes" are self-contained signed tokens, so this works fine on Vercel's stateless serverless functions.
Tools exposed
| Tool | What it does |
|---|---|
shiprocket_cost_summary |
Actual Shiprocket freight cost for yesterday / last_7_days / last_30_days / mtd |
shiprocket_raw_order_sample |
Diagnostic — 3 raw orders, to confirm the charge field name |
delhivery_cost_estimate |
Estimated Delhivery cost for a shipment list you supply |
combined_cost_summary |
Both carriers combined into one INR total |
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.