How can the last indexed fetch time be returned with Brave context?

Definition: Brave LLM Context returns fetched_content_timestamp as Unix seconds for the last fetch of a referenced page by the index. Add enable_source_metadata=true to request this field in the sources[url] metadata object.

Key facts

• Brave LLM Context returns extracted page chunks and source metadata for an AI application, rather than the short snippets used by its traditional Web Search results page. Brave's LLM Context documentation describes the response as metadata keyed by URL.
• enable_source_metadata is a Boolean query parameter whose default is false. When enabled, it adds site_name, favicon, thumbnail, description and fetched_content_timestamp to each sources[url] entry.
• fetched_content_timestamp is an integer containing Unix seconds of the index's last fetch of the page. Brave's API reference says the field requires enable_source_metadata.
• The API reference describes the option as source metadata enrichment in the response's sources attribute and repeats the default value false.
• Brave's 31 July 2026 changelog entry lists fetched_content_timestamp among the fields added by enable_source_metadata.

What does the timestamp describe?

Brave's response format documentation defines sources as metadata for all referenced URLs, keyed by URL. A source entry includes the page title, hostname and age array. The age array represents a date reported by the page in four fixed positions, while fetched_content_timestamp records when the index last fetched the page.

The two fields answer different questions. sources[url].age describes a date associated with the page. sources[url].fetched_content_timestamp describes an index event and uses Unix seconds. The documentation does not state a refresh interval or a fallback value for every URL.

How do you request the field?

The Brave API reference lists enable_source_metadata as an optional Boolean query parameter. Set it to true on the LLM Context endpoint and keep the API key in the X-Subscription-Token header.

curl -X GET "https://api.search.brave.com/res/v1/llm/context?q=tallest+mountains+in+the+world&enable_source_metadata=true" \
  -H "X-Subscription-Token: <YOUR_API_KEY>"

The response keeps the field under the URL-keyed source object:

sources[url].fetched_content_timestamp

Brave's LLM Context documentation also states that POST accepts the same query parameters as a JSON request body with Content-Type: application/json. An integration can therefore use the same Boolean setting in either request style.

How should you interpret the returned value?

Treat the value as an integer timestamp from Brave's index. Convert it to the display time needed by your logs or interface, while retaining the original value for comparisons between responses. Do not substitute the page's published or modified date: Brave documents those dates through the separate age field and defines this field as the index's last fetch time.

The timestamp describes the copy used by Brave's index at the recorded fetch event. It does not report when a website owner last edited the page, and the documentation gives no refresh schedule for the index.

Related terms

• How does Brave LLM Context differ from ordinary web results?
• How can low-relevance chunks be filtered from Brave LLM Context?
• Brave Search API

FAQ

Is fetched_content_timestamp returned by default?

No. Brave documents enable_source_metadata with a default of false. Set the parameter to true when the response needs the extra source fields.

Is the value an ISO 8601 date?

No. The field is an integer expressed in Unix seconds. The age array has an ISO 8601 rendering among its four page-date positions, which is a separate value.

Can a POST request include the option?

Yes. Brave documents that POST accepts the same query parameters as a JSON request body, so enable_source_metadata can be included there with the other request parameters.

Sources

1. Brave Search API, "LLM Context: Web search for AI agents and chatbots", 31 July 2026
2. Brave Search API, "LLM Context Web search API for AI agents API Reference", undated

---

Check how a page reads to answer engines. Open the GEO score analyzer