> For clean Markdown of any page, append .md to the page URL.
> For a complete documentation index, see https://docs.beatapi.io/llms.txt.
> For AI client integration (Claude Code, Cursor, etc.), connect to the MCP server at https://docs.beatapi.io/_mcp/server.

# Build a Channels video inventory

> Export a public Channels account's videos page by page with wechat_channels.v2.fetch_user_videos, then select cases from the CSV inventory.

To export a public Channels video inventory, fetch the account's video pages with `wechat_channels.v2.fetch_user_videos` and the account's `v2_…@finder` ID, then export the saved responses to CSV with the video inventory exporter. For example, to find a museum's exhibition guides from last month, list titles, publication dates and video IDs first, then shortlist the videos about reservations and routes. To look for one topic, [search within the account](/guides/wechat-channels-channel-search) instead.

## Confirm the account and save its video pages

1. Reuse a verified `v2_…@finder` username. If you have an `sph…` ID or a share link, confirm the account with the [Channels identifier workflow](/faq/wechat-channels-identifiers) first.
2. Send the request below to `POST https://api.beatapi.io/v1/social-data/call` with BeatAPI Bearer authentication, replacing the example username.

```json
{"action":"wechat_channels.v2.fetch_user_videos","params":{"username":"v2_123abc@finder","last_buffer":"","raw":false}}
```

3. Read the page's videos from `data.videos`.
4. If `data.up_continue` indicates more videos and a new `data.last_buffer` is returned, keep the same account and mode and pass the new `data.last_buffer` value back as `last_buffer`.
5. For each page, save the request, response, collection time and request ID.
6. Stop at a repeated cursor, an error or your page limit, and record the last successful page and the stop reason.

## Select useful cases from the inventory

Run the [video inventory exporter](/faq/wechat-channels-search-pagination) on the saved pages. Filter `selected-videos.csv` to the target month and record a selection reason by topic, such as reservations, arrival or visiting routes. For each selected video, keep the numeric video ID, the returned caption, the publication time and the source. Leave missing counts blank. When you pick cases by likes or shares, state that the comparison covers this run's sample, check the publication time and watch the actual video.

Before reporting scope, read the page count and stop reason in `coverage.json`. At the page limit, report the pages reviewed and the dates observed. At the end of the returned list, report that this run exhausted the available list. When a topic did not appear, report the topic together with the scope you checked, so the next run can extend the search.

For example, if the earliest video on page one is from the fifth of this month, record "checked one page; observed dates have not reached last month" and continue within your page budget. For play, audience or campaign reports on your own account, start in [Channels Assistant](/guides/wechat-channels-data-access).

For example, if three videos show the walk from the subway to the museum entrance, select one as a reference and read the selected video's [comments](/guides/wechat-channels-comment-review) for route questions the directions left open. Plan requests by page with the current [pricing](https://beatapi.io/pricing).