---
title: "Tool Result"
slug: "tool-result"
category: "glossary"
tags: ["glossary", "ai-agents", "llm", "tool-use", "agent-loop", "function-calling"]
status: "stable"
last_updated: 2026-05-14
summary: "A tool result is the structured response injected into the LLM's context after a tool call, providing the observation needed to continue reasoning."
related:
  [
    "[[glossary/tool-call]]",
    "[[glossary/tool-use]]",
    "[[glossary/agent-loop]]",
    "[[glossary/function-calling]]",
    "[[glossary/system-message]]",
    "[[glossary/hallucination]]",
  ]
---

## Overview

This page is the atomic definition. The tool call that precedes this is defined at [[glossary/tool-call]]. Full tool-use patterns live at [[glossary/tool-use]].

## Definition

A tool result is the content returned to the LLM after a [[glossary/tool-call]] has been executed by the runtime. In the Anthropic API, when the model emits a `tool_use` block, the caller executes the tool and sends back a `tool_result` block with the same `tool_use_id`. The model then receives this result as a new observation in its context and decides whether to call another tool, revise its plan, or emit a final response. The structure of tool results matters: verbose or unstructured results consume context window and may confuse the model. Best practice is to return the minimum information the model needs, structured as JSON or a short text summary. Error results should include a clear error code and message so the model can decide whether to retry or escalate. Tool results in the [[glossary/agent-loop]] are the primary mechanism through which the model observes the world; their quality directly affects the model's reasoning quality.

## When it applies

Handle tool result formatting explicitly in every tool implementation. Truncate large outputs (e.g., file reads beyond a few kilobytes) to a summary plus a signal that the full content is available via a follow-up call. Never return raw stack traces or multi-megabyte blobs as tool results.

## Example

Tool call: `read_file(path="/src/app.py")`. The runtime reads the file and returns:
```json
{"tool_use_id": "tu_abc", "type": "tool_result", "content": "# app.py (truncated to 500 lines)\n...file content..."}
```
The model uses this observation to decide the next step: call a code editing tool or ask for clarification.

## Related concepts

- [[glossary/tool-call]] - the model's request that generates this result.
- [[glossary/tool-use]] - the broader practice of equipping LLMs with external tools.
- [[glossary/agent-loop]] - tool results are the observation step in each loop iteration.
- [[glossary/function-calling]] - the equivalent mechanism in OpenAI's API.
- [[glossary/hallucination]] - a model that hallucinates tool results skips the execution step, breaking the loop.

## Citing this term

> See [[glossary/tool-result|Tool Result]] (llmbestpractices.com/glossary/tool-result).

## Related

- [[glossary/tool-call]]
- [[glossary/tool-use]]
- [[glossary/agent-loop]]
- [[glossary/function-calling]]
- [[glossary/system-message]]
- [[glossary/hallucination]]