The ability of an AI agent to interact with external tools and APIs — reading data, calling functions, and taking actions in the real world.
An LLM without tools can only generate text. An LLM with tools can check your calendar, query a database, send a message, or deploy code. Tool use is what transforms a chatbot into an agent.
In OpenClaw, tool use happens through MCP (Model Context Protocol). The agent discovers available tools, decides which ones to use, and calls them with appropriate parameters. The tool returns results that the agent incorporates into its reasoning.
Every tool call is a potential security surface. An agent with database access could drop a table. An agent with email access could send to wrong recipients.
Tool use is the defining feature that separates agents from chatbots. It is also the biggest security risk. Every tool the agent can access is an attack surface that needs monitoring and control.
Clawctl provides 200+ pre-configured tool integrations via MCP. Every tool call is logged in the audit trail. Risky tool actions require human approval. Egress filtering ensures tools only reach approved external domains.
Try Clawctl — 60 Second DeployAny tool with an MCP server. Clawctl provides 200+ pre-built integrations including GitHub, Slack, Stripe, Google Workspace, and more.
Yes. Build a custom MCP server for any API or tool and connect it to your OpenClaw agent.
Configure allowed tools in your openclaw.json. Clawctl adds approval workflows for risky tool operations.
MCP (Model Context Protocol)
An open protocol that lets AI agents connect to external tools and data sources through a standardized interface.
Function Calling
A capability of LLMs to generate structured function calls instead of plain text, enabling agents to interact with APIs and tools programmatically.
Egress Filtering
Network-level control that restricts which external domains an AI agent can communicate with, preventing data exfiltration.
Approval Workflow
A process where risky agent actions are paused and routed to a human for review before execution.