> ## Documentation Index
> Fetch the complete documentation index at: https://docs.usequotable.com/llms.txt
> Use this file to discover all available pages before exploring further.

# Log file import

> Upload a CSV, Common or Combined Log Format, or NDJSON access log to see recent AI agent traffic without configuring a live connector.

Import a log file when you want a quick look at recent AI agent traffic, or to fill the days before a live connector was set up. The file is read during the upload and then discarded: it's never stored.

> [!WARNING]
>
> Only requests from the **last seven days** are kept. Older rows are dropped as `timestamp_out_of_range`. File import is for diagnostics and short backfills, not for loading history.

## Limits

| Limit | Value |
| - | - |
| File size | Set by your plan. See [Plans and limits](https://docs.usequotable.com/workspace/plans-and-limits.md#brand-plans) |
| Lines per file | 200,000 |
| Imports | 20 per hour per workspace |
| Accepted events | Count toward the source's daily limit and the workspace's monthly allowance |

Upload the plain, uncompressed text file.

## Upload a file

**1. Create an import source**

In **Agent Analytics**, select **Connect source**, choose **Log file import** and select **Generate credentials**. The upload area appears in the same step.

**2. Choose the file**

Select a `.csv`, `.log`, `.txt`, `.ndjson` or `.jsonl` file and select **Import**.

**3. Read the receipt**

The import runs in the background. Select **Refresh** to update its status (`queued`, `running`, `done` or `failed`). When it's done, the receipt shows rows read, agent events accepted, duplicates and rows dropped.

Imported events are labelled as file imports, so they never affect the health of your live sources.

## Formats

The format is detected from the file name and its first line.

### CSV

A header row is required. Column names are matched case-insensitively against these aliases; other columns are ignored, so an export with extra columns is safe to upload.

| Field | Accepted column names |
| - | - |
| Time | `timestamp`, `datetime`, `occurred_at`, `@timestamp`, or `date` plus `time` |
| Host | `host`, `hostname`, `cs(host)`, `x-host-header` |
| Path | `path`, `uri`, `url`, `request_path`, `cs-uri-stem` |
| Method | `method`, `request_method`, `cs-method` |
| Status | `status`, `status_code`, `sc-status` |
| Bytes | `bytes`, `size`, `response_bytes`, `sc-bytes` |
| Duration | `duration_ms`, `latency_ms`, or `time-taken` in seconds |
| Country | `country`, `country_code`, `c-country` |
| Referrer | `referer`, `referrer`, `cs(referer)` |
| User agent | `user_agent`, `useragent`, `user-agent`, `ua`, `cs(user-agent)` |
| Client IP | `ip`, `client_ip`, `remote_addr`, `c-ip` |
| Cache status | `cache_status`, `x-edge-result-type` |
| Request ID | `id`, `request_id`, `x-edge-request-id` |

Rows without a host use the source's domain.

### Common and Combined Log Format

Standard Apache and Nginx access logs. CLF has no host, so every line is attributed to the source's domain, and no duration or cache status is available. Use the Combined format: plain Common Log Format has no user agent, so no line can be recognized as an AI agent.

```text theme={"system"}
203.0.113.10 - - [01/Oct/2026:09:12:44 +0000] "GET /pricing HTTP/1.1" 200 18432 "https://chatgpt.com/" "Mozilla/5.0 (compatible; GPTBot/1.2; +https://openai.com/gptbot)"
```

### NDJSON

One JSON object per line (or a single JSON array), with the fields of the [Custom HTTP](https://docs.usequotable.com/agent-analytics/connectors/custom-http.md#record-fields) connector.

## Duplicates

Uploading the same file to the same source again is refused with "This file was already imported for this source". A file whose import failed can be uploaded again.

## Next steps

> [card] [Choose a live connector](https://docs.usequotable.com/agent-analytics/getting-started.md#choose-a-connector)
>
> Keep data flowing after the backfill.

> [card] [Custom HTTP](https://docs.usequotable.com/agent-analytics/connectors/custom-http.md)
>
> Stream records from your own pipeline.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.