Tool Use
Tool use is the capability of a language model to invoke external functions, APIs, or services as part of producing a response.
What is Tool Use?
Tool use turns an LLM from a text generator into a controller of external systems. The model is given a list of tools with names, parameters, and descriptions; at inference time it can choose to emit a structured tool call rather than free text. The host system executes the call, returns the result, and lets the model continue. Tool use is the foundation of agentic AI, retrieval grounded in private APIs, and any workflow where the model needs fresh or private data.
How does Tool Use apply to enterprise AI?
Enterprise tool use enables AI agents to query CRMs, search knowledge bases, look up policies, draft tickets, send emails, and trigger workflows. Each tool definition is a security boundary and must be scoped, logged, and rate-limited.
Related terms
- Function Calling - Function calling is the specific implementation of tool use where the language model emits a structured JSON object matching a function signature, which the host application then executes.
- Agentic AI - Agentic AI refers to systems that plan multi-step actions, call external tools, and operate with some autonomy toward a goal, rather than producing a single response to a single prompt.
- Guardrails - Guardrails are runtime checks placed around an AI system to constrain inputs, outputs, and tool calls within safety, compliance, and business policy.
- AI Audit Trail - An AI audit trail is the persistent, tamper-evident record of every input, output, tool call, model version, and decision an AI system has made, sufficient to reconstruct any past interaction.
External references
Need help applying Tool Use to your enterprise? Submit a short brief and we reply within one business day.