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

# Knowledge source settings, 200 articles per page, and MCP source updates

Availability rules on knowledge sources shipped over the last two weeks for [web import sources](/2026-09-28-web-import-availability-rules), [Knowledge API sources](/2026-09-30-knowledge-api-source-availability-rules) and [the sources that an Ada App creates](/2026-10-01-integration-source-availability-rules). This note covers the changes that came with that work and have no release note of their own.

## Rename and delete a Knowledge API source from the dashboard

The **Settings** window of a Knowledge API source now holds more than its availability rules.

* **Source name**: rename a source that you created through the Knowledge API. No source can take the name **Created in Ada**.
* **Delete source**: delete the source and all of its articles, with the same confirmation as a web import source.

Open the window from the **Sources** tab with the **Settings** button beside the source. See [Connect through the Knowledge API](/docs/knowledge/content-ingestion/knowledge-integration#kb-api).

## 200 articles per page

The article list on the **Knowledge** page can now show up to 200 articles per page. The largest page size was 100. Bulk actions such as **Set as inactive** and **Set availability** apply to every selected article on the page. A cleanup of thousands of articles takes half as many pages. See [Article management](/docs/knowledge/article-management#enable-disable-articles).

## Knowledge sources over the MCP server

[`edit_agent_config`](/mcp/tools/edit-agent-config) and [`list_entities`](/mcp/tools/list-entities) now cover source settings and availability rules.

* `edit_agent_config` with `entity_type='knowledge_source'` accepts `availability_rules` on `create`, creates a website import when the request passes `start_urls` and an optional `sync_frequency`, and offers a new `update` operation. Depending on the source type, `update` changes the name, the sync frequency or the availability rules.
* `list_entities` with `entity_type='knowledge_sources'` returns each source's `availability_rules` at full detail.
* `list_entities` with `entity_type='variables'` lists every variable that an availability rule can reference, including `channel`, `chatter_id` and `end_user_id` and the `sensitive` and `sensitive_meta` scopes, so an assistant can find the variable ID a rule needs.