Skip to main content
Tools transform your Agent from a reasoning engine into an action-taker. Each tool is an external API endpoint that the Agent can call to interact with the world.
For a complete reference of all available actions and how they’re used across Cargo, see the Actions overview.
Tools configuration interface

​
How tools work

When an Agent receives a task, it:
  1. Analyzes the goal and available tools
  2. Selects the appropriate tool(s) based on their descriptions
  3. Executes the tool with the right inputs
  4. Uses the output to continue reasoning or complete the task
The tool’s description is critical. The Agent reads this description to decide when and how to use the tool. Write descriptions that clearly explain what the tool does and when to use it.

​
Configuring a tool

​
Description (The AI Contract)

This plain-language description tells the Agent:
  • What the tool does
  • When to use it
  • What inputs it expects
Good example: 
"Enriches a company using its domain. Returns firmographic data including
employee count, industry, funding, and tech stack. Use this when you need
company information for qualification or research."
Poor example: 
"Gets company data"

​
Input configuration

Control how the Agent provides inputs to the tool:
ModeBehaviorUse when
Let agent decideThe Agent autonomously determines inputs from contextInputs are straightforward and available in the conversation
Let me decideYou explicitly define the input structureInputs require specific formatting or come from fixed sources

​
Bulk execution

Enable this when the Agent should apply the tool across multiple records in a single run. Useful for batch operations like:
  • Enriching a list of companies
  • Updating multiple CRM records
  • Sending notifications to several recipients

​
Common tool types

CategoryExamplesTypical use
EnrichmentClearbit, Apollo, FullEnrichGather company or contact data
CRMSalesforce, HubSpot, AttioCreate/update records, log activities
CommunicationSlack, Email, OutreachSend messages, add to sequences
Data lookupInternal APIs, SQL queriesRetrieve custom data from your systems

​
MCP Servers

Cargo Agents support the Model Context Protocol (MCP)—an open standard for connecting AI models to external tools and data sources. You can connect your Agent to any MCP-compatible server, giving it instant access to a wide ecosystem of pre-built integrations. Simply provide the server URL, and Cargo will automatically discover and register the available tools. For example, you can connect your Agent to:
  • Notion — Search pages, create databases, and manage workspace content
  • HeyReach — Automate LinkedIn outreach and manage campaigns
  • HubSpot — Access CRM data, manage contacts, and trigger workflows

​
Best practices

Write clear descriptions

The Agent relies on descriptions to choose tools. Be specific about what the tool does and what it returns.

Name tools descriptively

Use names like Enrich-Company or Slack-Notify-Rep instead of generic names like Tool1.

Handle failures gracefully

In your Agent’s instructions, specify what to do if a tool call fails (retry, skip, report error).

Limit scope

Only give Agents access to the tools they need. Fewer tools = faster, more accurate tool selection.

​
Next steps

All actions reference

Complete reference for all action types available in Cargo.

Using actions everywhere

Learn how to leverage actions across tools, plays, agents, and APIs.