# Tool execution logs (/docs/observability/logs)

The Logs API returns **individual tool execution events** — one record per tool call. Use it to debug failures, inspect request/response payloads, and trace specific user activity. For aggregated counts (how many tool calls happened), use the [Usage API](/docs/observability/usage) instead.

All endpoints in this section require a **project API key** (`x-api-key`) or a valid session cookie.

# List logs

```bash
curl -X POST https://backend.composio.dev/api/v3.1/logs/tool_execution \
  -H "x-api-key: YOUR_PROJECT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "limit": 20,
    "time_range": {
      "from": 1744848000000,
      "to": 1744934400000
    },
    "filters": [
      { "field": "toolkit_slug", "operator": "==", "value": "gmail" },
      { "field": "status", "operator": "==", "value": "failed" }
    ]
  }'
```

The response contains a page of log entries and a `next_cursor`:

```json
{
  "logs": [
    {
      "id": "log_-jRTWClpBoVo",
      "timestamp": "2026-04-17T10:25:00.000Z",
      "type": "tool.execution",
      "status": "failed",
      "level": "error",
      "message": "GMAIL_SEND_EMAIL failed: invalid recipient",
      "metadata": { /* tool, toolkit, user_id, connected_account_id, ... */ },
      "metrics": { "duration_ms": 202 },
      "parent": null
    }
  ],
  "next_cursor": "eyJwYWdlIjoyfQ=="
}
```

Pass `next_cursor` back as `cursor` on the next request to paginate. When `next_cursor` is `null`, you've reached the end.

## Filter fields

Pass one or more filters in the `filters` array. Filters are **AND**-combined.

| Field                  | What it matches                                             |
| ---------------------- | ----------------------------------------------------------- |
| `tool_slug`            | The specific tool that was called (e.g. `GMAIL_SEND_EMAIL`) |
| `toolkit_slug`         | The toolkit (e.g. `gmail`, `slack`, `github`)               |
| `connected_account_id` | The connected account used for the call                     |
| `auth_config_id`       | The auth config (integration) behind the connected account  |
| `status`               | `success` or `failed`                                       |
| `user_id`              | Entity that initiated the call                              |
| `session_id`           | Tool router session, if routed through a session            |
| `sandbox_id`           | Sandbox the call ran in, if applicable                      |
| `request_id`           | Request ID (useful for correlating with your own logs)      |
| `log_id`               | Exact log ID (equivalent to the detail endpoint)            |

## Operators

| Operator       | Meaning                  |
| -------------- | ------------------------ |
| `==`           | Exact match              |
| `!=`           | Not equal                |
| `contains`     | Substring match          |
| `not_contains` | Substring does not match |

## Parameters

| Field             | Type           | Default | Notes                                          |
| ----------------- | -------------- | ------- | ---------------------------------------------- |
| `limit`           | number         | `20`    | Max 100                                        |
| `cursor`          | string \| null | `null`  | Opaque pagination token from previous response |
| `filters`         | array          | `[]`    | AND-combined                                   |
| `time_range.from` | number         | —       | Epoch milliseconds                             |
| `time_range.to`   | number         | —       | Epoch milliseconds                             |

# Get a single log

Fetch one log by ID to get the **full** payload, including request/response bodies, timing breakdowns, and source metadata:

```bash
curl https://backend.composio.dev/api/v3.1/logs/tool_execution/log_-jRTWClpBoVo \
  -H "x-api-key: YOUR_PROJECT_API_KEY"
```

The detail response includes everything from the list shape plus:

* `timings` — `start_time` and `end_time` in epoch ms
* `context` — `session_id`, `trace_id`, `request_id`
* `source` — `host` (e.g. `mcp`, `sdk`, `api`), `framework`, `language`
* `data` — the full request payload and response body

This is the endpoint to call when you need to reconstruct *exactly* what happened — e.g. when debugging a 500 from a user report.

# Recipes

## Find failed Gmail tool calls in the last hour

```bash
NOW=$(date +%s)000
HOUR_AGO=$(( $(date +%s) - 3600 ))000
curl -X POST https://backend.composio.dev/api/v3.1/logs/tool_execution \
  -H "x-api-key: YOUR_PROJECT_API_KEY" \
  -H "Content-Type: application/json" \
  -d "{
    \"time_range\": { \"from\": ${HOUR_AGO}, \"to\": ${NOW} },
    \"filters\": [
      { \"field\": \"toolkit_slug\", \"operator\": \"==\", \"value\": \"gmail\" },
      { \"field\": \"status\", \"operator\": \"==\", \"value\": \"failed\" }
    ]
  }"
```

## Get failures for a specific user

```bash
curl -X POST https://backend.composio.dev/api/v3.1/logs/tool_execution \
  -H "x-api-key: YOUR_PROJECT_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "filters": [
      { "field": "user_id", "operator": "==", "value": "user_abc123" },
      { "field": "status", "operator": "==", "value": "failed" }
    ]
  }'
```

## Fetch a single log's full request/response

```bash
curl https://backend.composio.dev/api/v3.1/logs/tool_execution/log_-jRTWClpBoVo \
  -H "x-api-key: YOUR_PROJECT_API_KEY"
```

# Reference

See the [Projects API reference](/reference/api-reference/projects) for the generated schema.

---

📚 **More documentation:** [View all docs](https://docs.composio.dev/llms.txt) | [Glossary](https://docs.composio.dev/llms.mdx/docs/glossary) | [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.


---

## Terminology Migration (old → current)

If you encounter these terms in error messages, old documentation, or user prompts, translate them to the current equivalents. **Do not use the old terms in generated code or explanations.**

| Old term (v1/v2) | Current term (v3) | In code |
|---|---|---|
| entity ID | user ID | `user_id` parameter |
| actions | tools | e.g., `GITHUB_CREATE_ISSUE` is a *tool* |
| apps / appType | toolkits | e.g., `github` is a *toolkit* |
| integration / integration ID | auth config / auth config ID | `auth_config_id` parameter |
| connection | connected account | `connected_accounts` namespace |
| ComposioToolSet / OpenAIToolSet | `Composio` class with a provider | `Composio(provider=...)` |
| toolset | provider | e.g., `OpenAIProvider` |

If a user says "entity ID", they mean `user_id`. If they say "integration", they mean "auth config". Always respond using the current terminology.