Office hours: Build your self-driving GTM engine·Oct 29, 9:00am PT, onlineRegister→
Cargo
  • PrimitivesData ModelsYour companies, contacts and deals, merged from every sourceContextWhat your team knows, as files every agent readsTerritoriesWho sells where, defined onceMailboxesMailboxes in your reps' names, warmed up and kept under limitsToolsThe actions your agents, plays and reps callAgentsAI that researches, decides and writes, reading your contextPlaysWorkflows that run on a trigger or a schedule, with retriesAppsInternal apps for your reps, hosted on Cargo
    Add-onsChrome ExtensionSell faster, right in your browserButtonTrigger AI agents inside your CRMFormTurn any tool into an embeddable formSlackTag Cargo in Slack. It does the GTM workGitHubPlan on every pull request, deploy on merge
    MoreProjectsDefine your entire GTM in codeCLIUse Cargo from your terminalMCPUse Cargo from Claude and ChatGPTDeveloper resourcesCargo API docs, OpenAPI spec, auth docs, webhooks, MCP serverAPI ReferenceExplore the Cargo APIDocumentationLearn how to use CargoIntegrationsConnected to your entire GTM stack
  • VeriffHow Veriff replaced MuleSoft with Cargo under NetSuite, Stripe and SalesforceAugmentHow Augment Code routes developer sign-ups to reps in under an hourWeights & BiasesHow Weights & Biases ships a new purchase signal in 2 days on CargoSee all
  • Pricing
  • Cookbooks
  • Blog
Sign in
×Cargo and Capy

Build your GTM in Capy.Cargo keeps it working.

One prompt connects Capy to Cargo. It writes your plays as code in your repo, and Cargo runs them with retries, rate limits and a trace for every run.

Set up in CapyRead the setup guide

One sentence in the agent you already use.

Capy~/my-gtm

Route every new HubSpot signup: enrich the company, score it against our ICP, and send good fits to the territory owner.

  • Readcontext/icp.md
  • Editedinfra/plays/inbound.ts+2
  • Rancargo-ai project plan && cargo-ai project deploy
    play:inbound created
Worked through 3 steps

Live. Every new signup is enriched, scored and routed to its owner, and each run is traced in Cargo.

A play in your repo, on Cargo's primitives, reviewed like any code.

import { definePlay } from "@cargo-ai/cdk";
 
import { contacts } from "../models/contacts";
import { routeFlow } from "../workflows/route";
 
// Runs once per new HubSpot contact, with retries and versioning
export const inbound = definePlay("inbound", {
  model: contacts,
  workflow: routeFlow,
  changeKinds: ["added"],
});

Every new contact, retried on failure, each run traced step by step.

SuccessAda Lovelace
2.41s6 spans601Nov 3rd, 2026 at 10:14:02
WaterfallSpans
Execution
0ms603ms1.21s1.81s2.41s
Ada Lovelace
2.41s
NorthwindEnrich company
640ms
82/100Scoring
310ms
United StatesSwitch
40ms
alice@acme.comAllocate
260ms
Owner updatedSalesforce Update owner
820ms
Message sentSlack Send message
240ms
~/the-toolkit/

All of Cargo, inside Capy

Capy builds with the same parts as your team does in the app. Cargo runs what it ships.

Data models

Companies, contacts and deals from every source, in one model.

Enrichment

Any field on any record, from the providers you choose, with the logic you write.

Tools

Typed actions like Find email, called from a play, the API or an agent.

Plays

Fire when a row changes or on a schedule. Retried on failure, traced end to end.

Routing

Territories, round-robin and capacity caps, applied as each lead arrives.

Context

Your ICP, personas and playbooks as files every play and agent reads.

~/what-you-ask/

What you ask Capy

Real jobs, from the request to the deploy.

Find out what broke, and fix it

Every run is traced, so the agent reads what happened instead of guessing. It finds the failing step, changes the play and ships the fix.

See every run traced
Capy~/my-gtm

Some signups were not routed last night. What happened?

  • Rancargo-ai orchestration run list --workflow-uuid 7c2d9e
    212 succeeded, 14 failed at enrich_company
Worked through 1 step

All 14 failed at the same step: the provider returned no domain for personal emails. I will skip enrichment for those and score them on the signup alone.

  • Editedinfra/plays/inbound.ts
  • Rancargo-ai project plan && cargo-ai project deploy
    play:inbound updated
Worked through 2 steps

Fixed. New signups with a personal email are scored and routed.

Start from a working project

Cookbooks are complete Cargo projects. The agent installs one, adapts it to your ICP and deploys it.

Browse cookbooks
Capy~/my-gtm

Install the account scoring cookbook and tier our companies against the ICP in context.

  • Rancargo-ai project add cookbook/account-scoring
  • Rancargo-ai project plan
    + create agent:account_tier_analyst+ create play:tier_accounts2 to create, 0 to update, 3 unchanged.
  • Rancargo-ai project deploy
    2 created, 0 updated, 3 unchanged.
Worked through 3 steps

Deployed. Every company gets a tier and the reason for it, hourly.

Review every change like code

Connect the repo once. Every pull request gets a Cargo plan, and every merge to main deploys.

Why GTM belongs in Git
Capy~/my-gtm

Connect this repo so every pull request gets a plan and every merge to main deploys. Then open a pull request for the inbound play.

  • Rancargo-ai project state connect --branch main
    { "outcome": "connected", "productionBranch": "main", "autoDeploy": true }
  • Opened a pull request
Worked through 2 steps

Cargo commented the plan on the pull request: 1 to create, 0 to update, 3 unchanged. Merging deploys it.

~/set-up/

One prompt sets it up

New accounts get free credits, no card.

  1. 1Paste the prompt into Capy.
  2. 2Approve each step: it installs the CLI and the Cargo skills, then signs you in with an emailed code.
  3. 3It scaffolds your GTM repo. Ask for your first play.

Prefer MCP? Connect the MCP server · CLI docs

Set up Capy
Install Cargo by following https://api.getcargo.io/INSTALL.md, then run npx @cargo-ai/cli@latest project init my-gtm and help me seed the repo for my company.
Paste it into Capy and approve each step.
$npm install -g @cargo-ai/cli cargo-ai login --email you@company.com npx @cargo-ai/cli@latest project init my-gtm cd my-gtm npx skills add getcargohq/cargo-skills --agent universal --yes git add .agents/skills && git commit -m "Add the Cargo skills"
Run them in order in your terminal.
~/traces/

Every run traced.Open any one.

Every run Capy triggers is traced. Open any run to see each step, what it returned and what it cost.

The Traces list, with Ada Lovelace's run open.

Traces
All triggers4/4 statusesLast 7 days
3,912 traces
Record
Trigger
Spans
User
Duration
Started at
Finished at
Credits
Ada Lovelace
Route inbound leads
6 spans60
2.41s
Nov 3rd, 2026 10:14:02
Nov 3rd, 2026 10:14:04
1
Northwind
Account researcher
3 spans30
AMAlice Martin
8.62s
Nov 3rd, 2026 10:11:40
Nov 3rd, 2026 10:11:49
3
Katherine Johnson
Route inbound leads
5 spans41
1.97s
Nov 3rd, 2026 10:09:13
Nov 3rd, 2026 10:09:15
1
Contoso
Enrich new accounts
4 spans40
3.18s
Nov 3rd, 2026 09:58:21
Nov 3rd, 2026 09:58:24
2
Grace Hopper
Route inbound leads
6 spans60
2.06s
Nov 3rd, 2026 09:51:07
Nov 3rd, 2026 09:51:09
1
Globex
Account researcher
4 spans40
MDMarc Dubois
11.3s
Nov 3rd, 2026 09:44:52
Nov 3rd, 2026 09:45:03
4
Alan Turing
Route inbound leads
6 spans60
2.28s
Nov 3rd, 2026 09:39:18
Nov 3rd, 2026 09:39:20
1
Initech
Enrich new accounts
4 spans31
4.05s
Nov 3rd, 2026 09:31:46
Nov 3rd, 2026 09:31:50
2
SuccessAda Lovelace
2.41s6 spans601Nov 3rd, 2026 at 10:14:02
WaterfallSpans
Execution
0ms603ms1.21s1.81s2.41s
Ada Lovelace
2.41s
NorthwindEnrich company
640ms
82/100Scoring
310ms
United StatesSwitch
40ms
alice@acme.comAllocate
260ms
Owner updatedSalesforce Update owner
820ms
Message sentSlack Send message
240ms

Cookbooks Capy can install

Working projects to start from, each one command away.

AS

Account scoring

Score every account against your ICP on a schedule.

CD

CRM deduplication

Deduplicate CRM accounts and contacts and keep them duplicate-free.

CE

CRM enrichment

Clean up and enrich a messy CRM.

ES

Expansion signals

Find which customers are ready to expand before renewal.

IQ

Inbound qualification

Qualify and route inbound demo requests from your website form.

NH

New hire detection

Get a Slack alert when a decision-maker joins a company in your ICP.

See all cookbooks
“Creating a tool: it works. Creating a play: it works. Creating an agent: it works too.”
Eliott M'Barali · GTM engineer working with Alma, building on Cargo from his coding agent

Frequently asked questions

Cargo and Capy: setup, plugins and MCP.

Yes. Capy drives Cargo through the Cargo CLI (cargo-ai) and the Cargo skills: it builds data models, context, tools, agents and plays as TypeScript in a repo you own, and Cargo runs them in production.
Paste one prompt into Capy: it installs the CLI from npm (@cargo-ai/cli), installs the Cargo skills, signs you in with an emailed code and scaffolds your GTM repo. To do it by hand, run npm install -g @cargo-ai/cli and sign in with cargo-ai login, scaffold the repo with cargo-ai project init, then from inside it run npx skills add getcargohq/cargo-skills --agent universal --yes and commit the .agents/skills folder.
Capy starts each session on a fresh machine, so the Cargo skills live in the repo: this writes them to .agents/skills. Commit them and every session has them.
Anything your GTM runs on: enrich and research accounts and contacts, score and route leads, write plays that run on a trigger or a schedule, build agents that read your context, and ship each change through a pull request that shows the plan before it deploys. Every run is traced in Cargo.
Yes. Every agent uses the same CLI, the same skills and the same workspace, so a teammate on a different agent works on the same project.
Yes. Cargo's MCP server at https://mcp.getcargo.io/mcp exposes the same workspace to any MCP client. For building and deploying a project, the CLI and skills go further.
~/agents/

Cargo works in every coding agent

Same CLI, skills and workspace, whichever agent you or a teammate use.

Terminal agents
  • Claude Code
  • Codex
  • Gemini CLI
  • Warp
  • OpenCode
  • Amp
  • Droid
Editors
  • Cursor
  • Windsurf
  • GitHub Copilot
  • Zed
  • Cline
  • Roo Code
  • Kiro
  • Junie
  • Antigravity
  • Augment
Cloud agents
  • Devin
  • CapyThis page
  • Replicas
  • Replit
Agent apps
  • Goose
  • Hermes Agent
  • OpenClaw
  • OpenHands
Your own agent
  • Any MCP client
All supported agents
~/ship-with-cargo/

Give your GTM workflows a runtime to ship on.

Run from terminal with Cargo CLI, then monitor results, governance, and execution in Cargo.

Start buildingRead the CLI docsStep-by-step setup

Want help with the first one? Book an onboarding call

Cargo
Backed byY CombinatorSOC 2 Type IICertified
LinkedinX

Product

Data ModelsContextTerritoriesMailboxesToolsAI AgentsPlaysAppsIntegrationsPricing

Add-ons

Chrome extensionButtonFormSlackGitHub

Build

DocsDeveloper resourcesCargo API docsCargo OpenAPI specCargo auth docsCargo webhooksCargo MCP serverProjectsClaude CodeCodexCursorAll coding agents

Resources

ReportsBlogCustomer storiesCookbooksEnrichmentRoutingSignalsScoringCRM hygieneAd audiencesGTM skillsGTM EngineeringGlossaryCompareCargo x ClayCargo x WorkatoCargo x n8nCargo x DeeplineCargo x DefaultCargo x LeanDataCargo x UnifyCargo x ValleyCargo x Gojiberry

Company

ManifestoCommunityEventsJobsPartners DirectoryPartner Program

Legal

SecurityTerms of servicePrivacy PolicyRefundsDPASupport
© 2026 Cargo
 ////////    ///////    ////////     ////////    ///////
///         ///   ///   ///   ///   ///         ///   ///
///         ///   ///   ///   ///   ///         ///   ///
///         /////////   ////////    ///  ////   ///   ///
///         ///   ///   /// ///     ///   ///   ///   ///
///         ///   ///   ///  ///    ///   ///   ///   ///
 ////////   ///   ///   ///   ///    ///////     ///////

$ npm install -g @cargo-ai/cli