cargo-mcp
Drive Cargo from its hosted MCP server at https://mcp.getcargo.io/mcp — connect a client, discover and price an action, run it over one record or a batch, poll it, and read workspace models, with no CLI install. Also when to call an MCP tool instead of shelling out to cargo-ai. Triggers: "connect Cargo to Claude Desktop", "add Cargo to ChatGPT", "Cargo MCP server", "mcp.getcargo.io", "use Cargo without installing anything", "which Cargo tool do I call", "searchactions", "executeactionbatch", "MCP server is showing the wrong workspace". Tools: whoami, searchactions, getactionschema, executeaction, executeactionbatch, getrun, querymodels. Skip when: you have a shell and the job is a workflow, a project deploy, warehouse SQL, or a mailbox — use the CLI skills; when publishing an MCP server out of your own workspace or attaching one to a Cargo agent — use cargo-ai.
- Installs
- 4.7K
- Last 24h
- +122
- Trending rank
- #1750
- Security
- Warn
Install
Works with every agent the skills CLI supports.
npx skills add https://github.com/getcargohq/cargo-skills --skill cargo-mcp- 01
Install
Run the command in your project. The CLI asks which agents to install for.
- 02
Describe the task
Ask your agent for something this skill covers. It loads cargo-mcp automatically when relevant.
- 03
Review the result
The agent follows the SKILL.md steps, scripts and references bundled with the skill.
Security audits
Independent scans of the skill's instructions and bundled files.
SKILL.md
The instructions your agent reads when the skill is triggered.
Cargo — the hosted MCP server
Cargo has two surfaces. The rest of this bundle documents the CLI. This one
documents https://mcp.getcargo.io/mcp, and, more usefully, when to reach for
which.
Three different things here are called MCP. This skill is the hosted server Cargo runs, which you point a client at. Publishing a curated server out of your own workspace (
ai mcp-server create, thencargo-ai mcpover stdio) and attaching somebody else's server to a Cargo agent (release update-draft --mcp-clients) are bothcargo-ai. Check which one the user means before answering: the words are identical and the answers share nothing.
Which surface
| The job | Surface |
|---|---|
| Run one action, or one action over many records | either; MCP if it is already connected |
| Find what Cargo can do, and what it costs | either (search_actions is the MCP half) |
| Read records off a model | either |
| One SELECT over models (aggregates, joins), 50 rows max | either (query_models is the MCP half) |
| Full SQL result sets, exports | CLI (cargo-storage) |
| Build or edit a multi-step workflow, tool, or play | CLI (cargo-orchestration) |
| GTM as Code: project plan and deploy | CLI (cargo-project) |
| Provision mailboxes, warm up, send | CLI (cargo-mailbox-management) |
| Segments, connectors, content libraries, alerts, hosting, billing admin | CLI |
| No shell at all (ChatGPT, Claude Desktop, claude.ai, n8n) | MCP, and say plainly what is out of reach |
The rule underneath the table: MCP is the runtime, the CLI is the platform. Seventeen tools cover discovering an action, running it, watching it finish, reading data back, and reading the workspace context. Everything that builds something reusable is CLI only. An agent holding both should prefer the CLI for anything the user will want to re-run or version, and MCP for one-shot execution inside a conversation.
When the job routes to the CLI, this is the whole bootstrap:
npm install -g @cargo-ai/cli
cargo-ai login --email you@company.com # emailed code, no browser; creates the account on first use
cargo-ai whoami # confirm the workspace before anything that spends
Connect
The endpoint is https://mcp.getcargo.io/mcp, Streamable HTTP. An
unauthenticated request returns 401 with a WWW-Authenticate challenge
carrying resource_metadata, so an OAuth-capable client discovers the
authorization server, registers, and prompts the user with no configuration
beyond the URL. A 401 on first connect is the handshake, not a fault.
claude mcp add --transport http cargo https://mcp.getcargo.io/mcp
Any client taking a JSON block (Claude Desktop, Cursor, a project .mcp.json):
{
"mcpServers": {
"cargo": {
"type": "http",
"url": "https://mcp.getcargo.io/mcp"
}
}
}
For CI, a headless agent, or a client with no OAuth, pass a workspace-scoped API token from Settings > API instead. Read it from the environment; never inline the value:
{
"mcpServers": {
"cargo": {
"type": "http",
"url": "https://mcp.getcargo.io/mcp",
"headers": { "Authorization": "Bearer ${CARGO_API_TOKEN}" }
}
}
}
This endpoint serves the platform tools only. A curated server a workspace
published with defineMcpServer is a different endpoint, at its own
/v1/ai/mcpServers/<uuid>/mcp URL. Every token sees the same platform tools
here; the token's permissions decide which calls succeed and which actions
search_actions returns.
The spine
whoami → which workspace am I in, how many credits
search_actions → find the action, and read its cost
get_action_schema → what inputs it takes
autocomplete_action → resolve a field needing a picked id (HubSpot object type, Slack channel)
execute_action │ one record
execute_action_batch │ many records
get_run / get_batch → poll while outcome is "executing"
For data: list_models → describe_model → query_models. For the workspace
context (read only): map_context, search_context, browse_context,
read_context. Alongside,
list_runs lists recent ad-hoc runs, and get_usage breaks the last 7 days of
credit spend down by integration.
Open every session with whoami. The token binds the session to exactly one
workspace and there is no flag to override it. A session pointed at the wrong
workspace returns plausible, confidently wrong reads: the models are real and
the records are real, they just belong to somebody else. Name the workspace back
to the user before acting on anything.
search_actions prices the work before you do it. Each result carries
credits[].cost beside the action object you pass verbatim to everything
downstream:
{
"name": "Enrich person & find email",
"credits": [{ "cost": 0.1, "type": "fixed" }],
"action": {
"kind": "connector",
"integrationSlug": "aiArk",
"actionSlug": "enrichPerson",
"connectorUuid": "7bb944ec-0254-44bc-b0e4-8a56378e80cf"
}
}
Pass that action object exactly as it comes, with no config key. Inputs
belong in data (single) or records (batch) — never in a config, which is a
node's configuration and has no meaning on a top-level action. Inputs
misplaced there are stripped without an error that names config: an action with
a required input is refused on the missing field, one without runs on nothing, so
an unexplained empty result is worth checking against this first. A misspelled key
in data or a batch record is refused before anything runs (invalidData,
Unknown field: <key>); tool actions are not checked.
get_action_schema takes the same pair: the action, plus an optional data for
the actions whose output depends on their inputs — a HubSpot object type or a
target sheet decides which fields come back. The CLI's
orchestration action get-output-schema behaves identically.
Four kind values come back: connector (a third-party integration), native
(a built-in platform operation), tool (a saved workflow in this workspace),
and agent (an AI agent in this workspace). Narrow a noisy catalog with the
kind and integrationSlug filters.
Three ways this goes wrong
Fanning out execute_action. One call per record is slower, bills more, and
leaves nothing to inspect afterwards. execute_action_batch takes the same
action plus a records array, produces one batch object, and a finished batch
returns one page of its output CSV in data.text (output columns prefixed
output_, 100 rows; while truncated
is true, call get_batch with the next offset). Read data.text, do not
fetch the URL. The tool description says never to loop
it: take that literally.
Spending before quoting. Run 10–20 records first (execute_action_batch
takes a limit that submits only the first N), report the observed
cost and hit rate, then quote the full record count and credit estimate
and let the user approve. Hit rates on people data run 40 to 70 percent, so cost
per usable row is not the sticker price and is not knowable without the
sample. Full discipline:
../cargo-gtm/references/cost-discipline.md.
Paging rows out of query_models. It runs one read-only SQL SELECT over
dataset.model references (take the exact reference from describe_model),
joins and aggregates allowed, and returns at most 50 rows (default 20). Answer
"how many", "grouped by", or "joined to" with the aggregate in SQL, never by
pulling rows and counting them yourself, which silently truncates at the cap. A
full result set is a CLI job (cargo-storage).
Anything that touches a person
The consent rules do not relax because the surface changed. A lawful basis, a
suppression check, and relevance to that person's job gate every step that
sources, enriches, or contacts someone. Bulk unsolicited messaging, purchased or
scraped lists, and consumer targeting are refused. The full text is
../cargo-gtm/references/acceptable-use.md;
where no sibling skill is installed, the paragraph above binds on its own.
Reporting back
Narrate and summarize; never paste raw JSON at a user. After a batch, give the record count, the hit rate, the credits actually spent, and the download, in that order.
FAQ
What does the cargo-mcp skill do?
Drive Cargo from its hosted MCP server at https://mcp.getcargo.io/mcp — connect a client, discover and price an action, run it over one record or a batch, poll it, and read workspace models, with no CLI install. Also when to call an MCP tool instead of shelling out to cargo-ai. Triggers: "connect Cargo to Claude Desktop", "add Cargo to ChatGPT", "Cargo MCP server", "mcp.getcargo.io", "use Cargo without installing anything", "which Cargo tool do I call", "searchactions", "executeactionbatch", "MCP server is showing the wrong workspace". Tools: whoami, searchactions, getactionschema, executeaction, executeactionbatch, getrun, querymodels. Skip when: you have a shell and the job is a workflow, a project deploy, warehouse SQL, or a mailbox — use the CLI skills; when publishing an MCP server out of your own workspace or attaching one to a Cargo agent — use cargo-ai.
How do I install cargo-mcp?
Run npx skills add https://github.com/getcargohq/cargo-skills --skill cargo-mcp in your project. The skills CLI detects the agents you use and copies the skill into the right folder. For Claude Code you can also place the skill folder at ~/.claude/skills/cargo-mcp/ (personal) or .claude/skills/cargo-mcp/ (project).
Which AI agents can use cargo-mcp?
Any agent that supports the open Agent Skills format (a folder with a SKILL.md file), including Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, Windsurf, Cline, OpenCode and more.
Is cargo-mcp safe to install?
Security partners have reviewed it: Gen Agent Trust Hub passed; Socket passed; Snyk flagged items to review. As with any skill, read the SKILL.md and bundled scripts before installing.
Who maintains cargo-mcp?
It is published by getcargohq in getcargohq/cargo-skills. It has 4,733 recorded installs.
More from getcargohq/cargo-skills
cargo-connection
getcargohq/cargo-skills
Connect Cargo to an external system and find out what it can do — authenticate connectors, browse the integration catalog, and resolve the connectorUuid and actionSlug a workflow node needs. Triggers: "connect my HubSpot", "is Salesforce connected", "what integrations do you support", "can Cargo talk to ", "what actions does have", "I need the connector UUID", "set up the API key for", "it is asking for credentials again", "why is this connector failing auth", "list my connectors". Integrations: amplemarket, amplitude, attio, bigQuery, calendly, cargoSignals, closecom, contrast, csv, customerio, dbt, emailBison, expandi, googleAds, googleCalendar, googleSheets, heyReach, http, hubspot, hubspotMcp, instantly, instantlyV2, intercom, jira, kitt, lemlist, lgm, linkedinAds, linkedinMatchedAudience, livestorm, manus, marketo, metaAds, metabase, microsoftTeams, mixpanel, monaco, netsuite, netsuiteSoap, nooks, notionMcp, octave, onesignal, outreach, pipedrive, postgresql,
cargo-gtm
getcargohq/cargo-skills
Do business-to-business go-to-market work on Cargo — research accounts and buying committees, enrich and verify B2B contact records from licensed data providers, score and qualify leads, draft permission-based outreach for the user's own sequencer, sync to CRM, and monitor buying signals. Consent basis, suppression lists, and volume limits gate every step that touches a person (references/acceptable-use.md); bulk unsolicited messaging, purchased or scraped lists, and consumer targeting are refused. Triggers: "build me a list of", "find 50 at ", "who works at", "find work emails for these accounts", "enrich this CSV", "verify these emails", "build a TAM", "who fits our ICP", "who actually buys from us", "what data points should we collect on accounts", "our outbound is reaching the wrong people", "score these leads", "write a first-touch email", "push these to my CRM", "who changed jobs", "who just raised funding", "companies using ", "who is hiring ", "fi
cargo-ai
getcargohq/cargo-skills
Build and configure AI agents inside Cargo — create an agent, choose its model and temperature, write its prompt, attach knowledge for retrieval (RAG), connect MCP tool servers, manage memories, and deploy releases. Triggers: "create an agent", "make an agent that", "give the agent our docs", "attach this knowledge base", "attach this library to the agent", "add resources to the agent release", "connect an MCP server", "expose our tools as an MCP server", "use Cargo from Claude Desktop or ChatGPT", "change the agent model", "what does the agent remember", "deploy the agent", "the agent is answering wrong". Skip when: uploading the knowledge files themselves — use cargo-content; sending the agent a message or running it over records — use cargo-orchestration.
cargo-workspace-management
getcargohq/cargo-skills
Administer a Cargo workspace and talk back to the Cargo team — invite and manage members, mint and rotate API tokens, organize plays, tools, and agents into folders, inspect roles, upload batch input files, and file reports. Triggers: "invite my teammate", "create an API token for CI", "who has access", "organize these into folders", "rotate that token", "upload this CSV for a batch" — and for feedback: "report this bug to Cargo", "send feedback to the Cargo team", "this CLI command is broken", "share this session with Cargo", "request a feature". Most commands need a token with admin access. Skip when: the question is about credits, plans, or invoices — use cargo-billing.
cargo-orchestration
getcargohq/cargo-skills
Make Cargo actually run something, or show what it would run — execute one connector action, run a multi-step workflow, trigger a batch across a whole segment or model, message an AI agent, build or edit a node graph, draw a workflow, tool or play as a diagram, and query the runtime tables (runs, batches, spans, records) with SQL. Triggers: "run this on all my contacts", "execute the action", "kick off a batch", "build a workflow", "schedule a play", "make it run every morning", "ask the agent", "show me the workflow", "what does this tool do", "visualize this play", "draw the graph", "explain this workflow", "how many runs failed today", "what is the output schema for this action", "add a step that". Skip when: explaining why a run misbehaved — use cargo-diagnostics; downloading result files — use cargo-analytics; committing the workflow as code — use cargo-project.
cargo-billing
getcargohq/cargo-skills
Understand what Cargo is costing — remaining credits, usage broken down by workflow, connector, or agent, subscription state, and invoice history. Triggers: "how many credits do I have left", "what did that cost", "why is my bill so high", "am I about to run out", "will this fit in our budget", "show me my invoices", "how much have I spent this month", "what plan am I on", "what do I get for free", "how many free credits", "can I afford this run", "add a card", "update my payment method", "why was my card declined". Needs a token with admin access. Skip when: attributing spend to specific nodes or cutting a play cost — use cargo-diagnostics.
Popular ai & agents skills
google-agents-cli-publish
google/agents-cli
This skill should be used when the user wants to "publish an agent", "publish my ADK agent", "register an agent with Gemini", "publish to Gemini" (formerly Gemini Enterprise), or needs guidance on the agents-cli publish gemini command. Also use when the user wants to "manage agents in Agent Registry", "list/update/delete registered agents", or "register an MCP server". Covers ADK vs A2A registration modes, programmatic and interactive usage, flag reference, auto-detection from deployment metadata, Agent Registry fleet management, and troubleshooting. Part of the agents-cli skills suite. Do NOT use for deployment (use google-agents-cli-deploy).
google-agents-cli-eval
google/agents-cli
This skill should be used when the user wants to "run an evaluation", "evaluate my agent", "evaluate my ADK agent", "write an eval dataset", "analyze eval failures", "compare eval results", "optimize agent", or needs guidance on the Gemini platform eval methodology and the Quality Flywheel. Covers eval metrics, dataset schema, LLM-as-judge scoring, and common failure causes. Applies to any agents-cli project, whatever framework the agent is written in. Do NOT use for agent API code patterns (ADK: use google-agents-cli-adk-code), deployment (use google-agents-cli-deploy), or project scaffolding (use google-agents-cli-scaffold).
skill-creator
anthropics/skills
Create new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.
codex-pet
prime-skills/runcomfy-agent-skills
Codex Pet generator on RunComfy. Build a Codex-compatible Codex Pet spritesheet.webp + pet.json from a single reference image, drop it into ${CODEXHOME:-$HOME/.codex}/pets/ / and Codex picks it up as a custom Codex Pet next to the 8 built-ins. This skill produces the exact Codex Pet atlas Codex expects (1536x1872 PNG/WebP, 8 cols x 9 rows, 192x208 cells, 9 animation states — idle, running-right, running-left, waving, jumping, failed, waiting, running, review). Calls OpenAI GPT Image 2 edit ONCE via the local RunComfy CLI as runcomfy run openai/gpt-image-2/edit to produce a canonical Codex Pet pose, then assembles all 9 animation rows programmatically with ImageMagick micro-transforms — no Codex Pro, no $imagegen, no OPENAIAPIKEY required, only RUNCOMFYTOKEN. Triggers on "codex pet", "create codex pet", "make codex pet", "hatch codex pet", "/hatch image", "desktop pet codex", "codex pets", "spritesheet.webp", or any explicit ask to build a custom pet for OpenAI Codex.
full-output-enforcement
leonxlnx/taste-skill
Overrides default LLM truncation behavior. Enforces complete code generation, bans placeholder patterns, and handles token-limit splits cleanly. Apply to any task requiring exhaustive, unabridged output.
executing-plans
obra/superpowers
Use when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available