# Tools and toolkits (/docs/tools-and-toolkits)

Composio offers 1000+ toolkits, but loading all the tools into context would overwhelm your agent. Instead, your agent has access to 5 meta tools that discover, authenticate, and execute the right tools at runtime.

# Meta tools

When you create a session, your agent gets these 5 meta tools:

| Meta tool                     | What it does                                               |
| ----------------------------- | ---------------------------------------------------------- |
| `COMPOSIO_SEARCH_TOOLS`       | Discover relevant tools across 1000+ apps                  |
| `COMPOSIO_MANAGE_CONNECTIONS` | Handle OAuth and API key authentication                    |
| `COMPOSIO_MULTI_EXECUTE_TOOL` | Execute up to 20 tools in parallel                         |
| `COMPOSIO_REMOTE_WORKBENCH`   | Run Python code in a [persistent sandbox](/docs/workbench) |
| `COMPOSIO_REMOTE_BASH_TOOL`   | Execute bash commands for file and data processing         |

Meta tool calls in a session are correlated using a `session_id`, allowing them to share context. The tools can also store useful information (like IDs and relationships discovered during execution) in memory for subsequent calls.

## How it works

```
User: "Create a GitHub issue for this bug"
    ↓
1. Agent calls COMPOSIO_SEARCH_TOOLS
   → Returns GITHUB_CREATE_ISSUE with input schema
   → Returns connection status: "not connected"
   → Returns execution plan and tips
    ↓
2. Agent calls COMPOSIO_MANAGE_CONNECTIONS (because not connected)
   → Returns auth link for GitHub
   → User clicks link and authenticates
    ↓
3. Agent calls COMPOSIO_MULTI_EXECUTE_TOOL
   → Executes GITHUB_CREATE_ISSUE with arguments
   → Returns the created issue details
    ↓
Done. (For large results, agent can use REMOTE_WORKBENCH to process)
```

## What SEARCH\_TOOLS returns

`COMPOSIO_SEARCH_TOOLS` returns:

* **Tools with schemas** - Matching tools with their slugs, descriptions, and input parameters
* **Connection status** - Whether the user has already authenticated with each toolkit
* **Execution plan** - Recommended steps and common pitfalls for the task
* **Related tools** - Prerequisites, alternatives, and follow-up tools

## Processing large results

For most tasks, `COMPOSIO_MULTI_EXECUTE_TOOL` returns results directly. But when dealing with large responses or bulk operations, your agent uses the workbench tools:

* **`COMPOSIO_REMOTE_WORKBENCH`** - Run Python code in a [persistent sandbox](/docs/workbench). Use for bulk operations (e.g., labeling 100 emails), complex data transformations, or when results need further analysis with helper functions like `invoke_llm`.

* **`COMPOSIO_REMOTE_BASH_TOOL`** - Execute bash commands for simpler file operations and data extraction using tools like `jq`, `awk`, `sed`, and `grep`.

# Toolkits and tools

A **toolkit** is a collection of related tools for a service. For example, the `github` toolkit contains tools for creating issues, managing pull requests, and starring repositories.

A **tool** is an individual action your agent can execute. Each tool has an input schema (required and optional parameters) and an output schema (what it returns). Tools follow a `{TOOLKIT}_{ACTION}` naming pattern, like `GITHUB_CREATE_ISSUE`.

> If you know exactly which tools you need, you can [execute them directly](/docs/tools-direct/executing-tools) without meta tools.

# Authentication

Tools execute with the user's authenticated credentials. When a user connects their GitHub account, all GitHub tools run with their permissions.

If a tool requires authentication and the user hasn't connected yet, the agent can use `COMPOSIO_MANAGE_CONNECTIONS` to prompt them.

- [Authentication](/docs/authentication): 
Learn how Composio handles user authentication

# Useful links

- [Browse toolkits](/toolkits): 
Explore all available toolkits

  - [Fetching tools](/docs/toolkits/fetching-tools-and-toolkits): 
Browse the catalog and fetch tools for sessions

---

📚 **More documentation:** [View all docs](https://docs.composio.dev/llms.txt) | [Cookbooks](https://docs.composio.dev/llms.mdx/cookbooks) | [API Reference](https://docs.composio.dev/llms.mdx/reference)

---

# Composio SDK — Instructions for AI Code Generators

**Purpose:** Enforce only the **current** and **correct** patterns for integrating [Composio](https://composio.dev/) tools into AI agents.
**Scope:** All AI-generated code using Composio must follow these guardrails.

---

## 1. Correct Composio Integration

Composio supports two integration modes: **Native Tools** (with a provider package) and **MCP** (no provider package needed).

### Native Tools

```python
# ✅ CORRECT — Python (defaults to OpenAI)
from composio import Composio

composio = Composio()
session = composio.create(user_id="user_123")
tools = session.tools()
# Pass tools to your agent/LLM framework
```

```typescript
// ✅ CORRECT — TypeScript (defaults to OpenAI)
import { Composio } from "@composio/core";

const composio = new Composio();
const session = await composio.create("user_123");
const tools = await session.tools();
// Pass tools to your agent/LLM framework
```

For other providers, pass the provider explicitly. Provider packages follow the naming convention: `composio_<provider>` for Python, `@composio/<provider>` for TypeScript.

### MCP

Use `session.mcp.url` and `session.mcp.headers` with any MCP-compatible client (Claude Desktop, Cursor, OpenAI Agents, etc.). No provider package needed.

---

## 2. CRITICAL INSTRUCTIONS FOR AI MODELS

### 2.1 — ALWAYS DO THE FOLLOWING

1. **Use `composio.create(user_id)`** to create a session — this is the entry point for all Composio integrations.
2. **Use `session.tools()`** for native tool integration or **`session.mcp.url`** for MCP integration.
3. **Import** from the correct provider package (`composio_<provider>` for Python, `@composio/<provider>` for TypeScript).
4. **Pass `user_id` to `composio.create()`**, not to individual tool calls.

### 2.2 — NEVER DO THE FOLLOWING

1. **Do not** instruct users to manually create auth configs or connected accounts before running.
2. **Do not** import from `composio_openai` / `@composio/openai` when using OpenAI Agents SDK — use `composio_openai_agents` / `@composio/openai-agents` instead.

### 2.3 — DISCOURAGED (do not use unless user explicitly asks)

Do **not** generate `composio.tools.get()`, `composio.tools.execute()`, or `composio.provider.handle_tool_calls()` unless the user explicitly requests direct tool execution. These are a supported but **not recommended** low-level interface.