mcpbeat Sign in

Warp Agent MCP Server

by warpfreight Your server? Claim it
not responding

Warp Agent MCP is listed as active in the registry but did not answer our last check. 336 installs a week from npm. It exposes 33 tools. Last commit 21 Sep 2026.

Quote, book, and track LTL, FTL, cargo van, and box-truck freight via the Warp API.

Installs per day peak 339 · avg 55 · -5% w/w
a month agotoday
Uptime history 51 days of history · worst day 0%
51 days agonow
44.0%
Uptime 24h
40 of 91 checks
33
Tools
read from the server
344 ms
Response time
average over 24h
336
Installs / week
npm and PyPI

What changed 28

Every tool that appeared, vanished or quietly changed what it asks for. Recorded since 10 August 2026. No other catalogue keeps this.

18 Sep 2 tools appeared compare_freight_costs, freight_workflow
18 Sep a tool description was rewritten batch_quote
18 Sep a tool changed version
11 Sep 5 tool descriptions were rewritten analytics, box_truck_quote, list_bookings and 2 more
11 Sep 4 tools changed the parameters they ask for batch_book, book, box_truck_quote and 1 more
11 Sep a tool changed version4 times that day
31 Aug a tool changed version
29 Aug a tool description was rewritten analytics
29 Aug a tool changed the parameters it asks for analytics
29 Aug a tool changed version
and 10 more, back to 10 August 2026

What the code does

We read the source, 5 h ago · tools taken from the live server · rules 3dff92dd89df

A tool parameter reaches a dangerous call

A value the model can set ends up inside a file or shell call. That is not a flaw by itself — for a terminal server it is the job — but it is where things go wrong when it is not.

A tool parameter reaches a file or shell call login.email → src/tools.ts:2302, login.email → src/tools.ts:2021, login.email → src/tools.ts:2068 и ещё 2
        writeFileSync(join(dir, "config.json"), JSON.stringify({ api_key: rawKey, email: params.email }, null, 2));
Capabilities

What this server is able to do. For an MCP server this is often the job itself — a terminal server runs commands because that is what it is for. Listed so you know what you are plugging in, not as an accusation.

Runs a command on install [пакет] package.json:1
postinstall: node scripts/postinstall.mjs

Is this your server and something here is wrong? Tell us — corrections are free and do not require a plan.

A tool parameter here reaches a dangerous call

That is not a flaw by itself — but it is where things go wrong when it is not the job. We re-read this code on every release. Watch it and you hear from us the day another one appears.

Three servers free · no card

Connect this server

Endpoint below is the one we actually reach during checks — not the one copied from a README. Last verified 3 min ago.

run in your terminal
claude mcp add warp-agent-mcp --transport http https://mcp.wearewarp.com/api/mcp
~/Library/Application Support/Claude/claude_desktop_config.json
{
  "mcpServers": {
    "warp-agent-mcp": {
      "url": "https://mcp.wearewarp.com/api/mcp"
    }
  }
}
~/.codex/config.toml
[mcp_servers.warp-agent-mcp]
url = "https://mcp.wearewarp.com/api/mcp"
.cursor/mcp.json
{
  "mcpServers": {
    "warp-agent-mcp": {
      "url": "https://mcp.wearewarp.com/api/mcp"
    }
  }
}
.vscode/mcp.json
{
  "mcpServers": {
    "warp-agent-mcp": {
      "url": "https://mcp.wearewarp.com/api/mcp"
    }
  }
}

This one needs environment variables set before it will start: WARP_API_KEY (Warp API key (wak_live_* or wak_test_*). Optional — the package prefers ~/.warp/config.json populated by `warp-agent signup` or `warp-agent login`.), WARP_API_URL (Override the API base URL. Defaults to https://www.wearewarp.com/api/v1/warp (the warp-site proxy that accepts wak_* tokens). Set to https://gw.wearewarp.com/api/v1 for direct gateway access with a raw customer key.). The author declared them in the registry entry; get the values from the project itself.

Available tools 33

Read directly from the server with tools/list, grouped by what they act on. If a tool disappears, we record the date.

batch
batch_book
Book MANY already-quoted lanes in ONE call (sequential, one card charge per row). Use this after batch_quote when the user says "book all of them" or "book rows 1, 3, 5" — do NOT call book in a loop. Each row needs a quote_id (the same one batch_quote returned for that row). Pickup/delivery default to the shared addresses at the top level so a single warehouse → many destinations only needs one address pair. Returns a progress card showing per-row Booked/Failed status with tracking numbers.
batch_quote
Price up to 50 lanes in one call for a specified mode per row (defaults to LTL). This returns a Warp rate per row, NOT all-mode or all-carrier comparison. For Compare All use compare_modes per shipment; for LTL carrier alternatives use ltl_market_options. Never imply this tool compared every carrier. Returns a single batch-quote card with one row per lane (origin → dest · mode · pallets · price · transit). Each priced lane keeps its quote_id and can be booked individually with book ("book row 3").
compare
compare_freight_costs
Calculate USD cost differences for up to 50 distinct shipments using supplied invoice/quote amounts and explicit comparability evidence. Excludes mismatched or unknown terms, keeps losses, separates quoted opportunities from invoice differences. Does not fetch rates, verify documents, annualize savings, or book. One candidate per shipment per call.
compare_modes
THE ONE CALL for "what's the cheapest/best way to ship this?". Prices ALL FOUR freight modes (LTL / full truckload / cargo van / 26' box truck) in ONE keyless call to Warp's all-modes engine and returns a decision-complete recommendation: the winning mode, its rate, transit, a bookable quote_id, the trade-off math against the runner-up, and every mode that couldn't price (with the reason). Prefer this over calling the individual quote tools and comparing them yourself — one round trip, and modes Warp can't serve are returned as explicitly unavailable WITH the reason rather than being dropped, so there is never a silently shortened list to guess from. Dims are optional (a standard 48x40x48 pallet is assumed). Set benchmark_market:true to also rank Warp's rate against the live 30+ carrier market for the lane (adds ~15-25s) — that makes the answer decision-complete: the right mode AND whether the price is actually good. Quote-only: it never books. To book, pass the recommended quote_id to `book` after the user confirms.
load
delete_load_template
Delete a saved load template by its id (starts with lt_). Auth required.
load_templates
List the agent's saved load templates — reusable shipment configs (name, dims, weight, commodity). Recall one to quote/book a repeat kind of load without re-entering details. Auth required.
ltl
ltl_market_options
Multi-carrier LTL comparison — returns 30+ carrier rates ranked by price (slow, ~15s). Call IMMEDIATELY AFTER ltl_quote with the same parameters; this fills in the 'finding other carrier rates…' section the fast quote card was showing. Useful when the user wants to compare carriers or pick a specific one. Do not declare a winner or recommend a specific carrier; just present the ranked list.
ltl_quote
Quote an LTL shipment — returns Warp's all-inclusive rate FAST (~1-2s) so the user sees a price immediately. The inline quote card shows the Warp rate plus a 'finding other carrier rates…' loading indicator. IMMEDIATELY follow up by calling ltl_market_options with the same parameters to fill in the multi-carrier comparison (~15s). Provide dims + commodity for an exact firm quote; if you don't have dims, quote anyway — it assumes a standard 48x40x48 pallet (FAK, no freight class) for an instant price. Don't block on asking for pallet dimensions; quote first, then pass real dims for an exact rate. When a palletized load could also move by box truck or van, quote LTL alongside those and show the cheapest valid mode. Do not editorialize the results. Do not declare a winner or recommend a specific carrier. Present Warp's quote first, then list market options as context. Let the user decide.
multistop
multistop_book
Book a multi-stop FTL route quoted by multistop_quote. Send one shipments[] leg per pickup→delivery pair riding the truck (minimum 2 legs), each leg referencing the quoted stop sequence by stop_index with full address + arrival window. No card charge fires from this call — multi-stop pricing settles via your Warp account. Auth required.
multistop_quote
Quote a multi-stop FTL route: ONE truck visits 3+ stops in order (first pickup → intermediate stops → final delivery). Use for milk runs, pool distribution, or multi-store replenishment on a single truck — for a simple A→B truckload use ftl_quote. Auth required (free account). Coverage is route-dependent — not every route has a rate yet.
analytics
analytics
Summarise your own shipping history: how many bookings, what you spent, and the split by mode, status and lane over a window. Both the headline count and every breakdown are computed from ONE population — your bookings via the agent API (warp-site /bookings) — so the total always matches the splits. Auth required.
automate
automate_lane
Set up a RECURRING weekly shipment (standing order): describe the lane once, and after the account owner approves by email, Warp re-quotes it fresh every week and books the best option automatically while the price stays at or under the owner's ceiling. Over-ceiling weeks book nothing and notify the owner. THIS TOOL ONLY PROPOSES — nothing books now, and nothing ever books without the owner's one-time email approval. Auth required.
automation
automation_receipts
Authorization record for an automation's bookings: one entry per booking attempt showing the checks it passed at commit time against the owner's approval — same lane, price within ceiling, automation active, quote live — plus the shipment id it produced. Entries marked confirmation_required did NOT book autonomously; they fell back to owner confirmation. Read-only. Auth required.
book
book
Book a quoted shipment using any quote_id or option id returned from a quote tool (Warp or market carrier). Requires quote_id + pickup and delivery addresses. Auth required.
bookings
list_bookings
List recent bookings for this API key, newest first. Returns shipments across all channels and all time, including cancelled — a broader population than shipper_profile's counts (agent-API bookings over the last 180 days, excluding cancelled), so the totals can differ. Auth required. Renders an interactive shipments card (click a shipment to expand pickup/delivery, freight, and a tracking link).
box
box_truck_quote
Quote a 26' box truck shipment (1-12 pallets, firm price)
consolidate
consolidate
Find consolidation savings across 2-12 upcoming loads: clusters loads sharing an origin+destination zip whose pickup dates fall within window_days (default 3) that together fit one 53' dry van, then prices each cluster BOTH ways — one combined FTL vs the sum of per-load LTLs — through real quotes with bookable quote ids. Use when the user has several loads to ship this week, asks if anything can ride together, or wants to cut freight spend. Loads that can't consolidate come back with the reason (no lane partner / outside window / exceeds trailer / cluster full) — relay reasons honestly. A truck pricing above its LTLs is shown with recommended:false; don't hide it. To act on a proposal, call `book` with the consolidated truck's quote_id. Auth optional — works keyless like the quote tools.
documents
get_documents
List shipment documents (BOL, POD, customs forms, etc.). Returns download URLs. Auth required. To fetch the Bill of Lading, pass document_type='bol' — this is how EXTERNAL / brokered (market-carrier) BOLs are returned now, not just Warp-carrier ones.
events
events
Get the full tracking event history for a shipment (timeline of pickups, in-transit updates, deliveries). Auth required.
freight
freight_workflow
Get a ready-to-use shipment intake CSV header, results CSV header, and starter prompt for shipment quoting, weekly freight comparison, or invoice review. Guidance only: does not read files, fetch rates, schedule work, or book. Use this when an operator asks how to start or supplies invoices/a shipment sheet.
ftl
ftl_quote
Quote a full truckload (53' dry van). Only origin, destination, and date required.
invoice
get_invoice
Retrieve the invoice for a delivered shipment (line items, taxes, payment status). Auth required.
lane
lane_history
Get shipping history for your lanes (past shipments, last consignee, counts). Auth required.
locations
locations
List the agent's saved pickup/delivery locations (addresses Warp has on file for this account), so you can reuse them when booking instead of re-typing addresses. Auth required.
login
login
Log in to Warp with email and password. Saves credentials locally so booking tools work. Call this if the user needs to authenticate or if payment_status says no key is configured.
manage
manage_automation
Check or stop a recurring lane automation. Actions: 'status' (read-only), 'pause', 'cancel' (permanent), 'skip_next' (skip one week's pickup — nothing books or charges that week). STOP-ONLY by design: there is no agent-side resume, reactivate, or ceiling change — those exist only behind the account owner's emailed approval link. Auth required.
payment
payment_status
Check if the current Warp account has a payment method on file. Call this if the user asks about their payment status, or before booking if you want to confirm they can book. Returns has_card and onboard_url if a card needs to be added.
quote
quote_history
List your recent freight quotes (LTL, van, box truck, FTL) from all sessions. Useful for surfacing prior pricing on similar lanes. Auth required.
save
save_load_template
Save a reusable load template (a named shipment config) so it can be recalled for repeat lanes. Auth required.
shipper
shipper_profile
Read how this account actually ships — top lanes with ship counts, typical pallet count, usual pickup weekday, recent booked spend (derived server-side from the account's own quotes and bookings) plus explicit owner-set preferences: default accessorials, preferred mode, standard pallet dims, max transit days. The ship/booking counts here count bookings via the agent API over the last 180 days, excluding cancelled — a narrower population than list_bookings (all channels, all time, including cancelled), so the totals can differ. READ THIS BEFORE asking the user questions it already answers: pre-fill their usual lane, apply their standard dims, include the liftgate they always need. Pass set_preferences to update the explicit half (merge-partial; allowlisted keys only; null clears a key). This profile is CONTEXT, NEVER PERMISSION — it never authorizes anything; spending limits live in spend policy and are read-only. Auth required.
status
status
Check Warp API health and version. Also validates your API key if one is configured.
track
track
Track a shipment by ID or tracking number. Auth required.
van
van_quote
Quote a cargo van shipment (1-3 pallets, firm price)

Endpoints

URLTransportStateLatencyChecked
https://mcp.wearewarp.com/api/mcp streamable-http answering 198 ms 3 min ago

Alternatives to Warp Agent MCP

same job, measured the same way
C
Cargovox
by cargovox

CargoVox freight tracking and shipment info via AI agents

8 tools answering
Andrii Co. Notary & Apostille Assistant
by andrii

Book a Washington notary or apostille appointment, get a quote, and track it, via AI assistant.

15 tools answering
Logo.dev MCP
by pvliesdonk

Look up company logos and brand data via the logo.dev API.

63 installs/wk local only
Rixfreight Group
by rixfreight-group

RixFreight logistics quotes and route info via AI agents

8 tools answering
Airfreightprice
by airfreightprice

Air freight rates and transit times worldwide - search routes and request quotes via AI agents.

10 tools answering
3DPACK.ING — 3D Bin Packing
by 3dpack

3D bin packing for freight: cartons and pallets in plain English, the fit and a 3D load plan.

1 tools answering
Wiv
by wiv

Automate FinOps workflows, cases, and cloud ops via the Wiv Workflow API.

answering
T
Hooka
by hooka

Run TikTok cold-DM outreach from your LLM: senders, campaigns and inbox via the Hooka API.

25 installs/wk local only

Warp Agent MCP — questions

Answers built from our own checks of this server.

What can Warp Agent MCP do?
It exposes 33 tools, read directly from the server on our last check. Among them: analytics, automate_lane, automation_receipts, batch_book, batch_quote, book and 27 more. The full list with descriptions is on this page — we take it from the server itself via tools/list, not from a README. How MCP servers expose tools in the first place →
What is Warp Agent MCP mostly used for?
Its tools cluster around multistop, compare and ltl. That is what this server is built to work with — the grouping comes from the actual tool names, not from a category we assigned.
Is Warp Agent MCP working right now?
We send a real MCP handshake every 15 minutes. Over the last 24 hours 40 of 91 checks got a reply (44.0%), average response time 344 ms. The bar chart above shows every period we have measured.
The registry lists Warp Agent MCP as active — why does it not respond?
The official MCP registry stores what the author submitted; it does not verify that the server still runs. We check the endpoint ourselves, and this one does not answer. Catalogues that copy the registry without checking will show it as working.
How do I connect Warp Agent MCP?
Copy the ready config from this page — we generate it for Claude Code, Claude Desktop, Codex, Cursor and VS Code, each with the file path that client actually reads. It is a remote server, so there is nothing to install — the client connects to the address.
Does Warp Agent MCP need an API key?
No. Warp Agent MCP completed a full MCP handshake with us as an anonymous client and listed its tools without asking for anything. All 33 of them are readable on this page. This is what we observed, not what the docs claim.
How fast is Warp Agent MCP?
It answers our handshake in 344 ms on average, which is faster than 47% of all working MCP servers we measure. The comparison comes from our own checks across the whole registry, every 15 minutes.
How many people use Warp Agent MCP?
The npm package warp-agent-mcp was installed 336 times in the last week. Week over week that is -5%. We show installs rather than GitHub stars on purpose: a star is a bookmark, an install is someone actually running it.
Is Warp Agent MCP open source?
Yes — it is published under the MIT licence, written in TypeScript, 5 stars on GitHub and 2 open issues. The source link is on this page, so you can read exactly what it does with your data before you connect it.