# Brave Search LLM Context

> Brave Search LLM Context is a paid API for AI agents from Brave Search at brave.mpp.paywithlocus.com, paid per call via MPP, $0.035/call, status healthy (last checked 2026-10-12, last successful call 2026-10-12).

Fetches web search results from Brave's independent index and returns distilled, token-limited content optimized for LLM consumption

## Facts

- Endpoint: POST https://brave.mpp.paywithlocus.com/brave/llm-context?utm_source=zero.xyz
- Price: $0.035/call
- Payment: MPP
- Status: healthy
- Last checked: 2026-10-12
- Last successful call: 2026-10-12
- Success rate: 98% of calls made through Zero
- Rating: 3.8 / 5 from 5 reviews
- Activations on Zero: 63
- Provider: Brave Search
- Docs: https://api.search.brave.com/app/#/documentation
- Website: https://brave.com/search
- Tags: search, web-search, privacy, news, images
- Canonical page: https://www.zero.xyz/c/brave-search-llm-context-d4d573af
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_k4zCc9V7Se45USb0ZFHTb

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability brave-search-llm-context-d4d573af -d '<json body>'
```

Example prompt: Search Brave for 'quantum computing breakthroughs' using results from the past week, limit it to 10 URLs and 8192 tokens, and give me the content formatted for my LLM context window.

## When to prefer this

Use this endpoint when you need web search results pre-processed and condensed into LLM-ready text rather than raw HTML or structured JSON. It is ideal for retrieval-augmented generation (RAG) pipelines, AI assistants that need fresh web grounding, or any agent that wants privacy-first Brave Search results without having to scrape and parse pages itself. Prefer it over generic web search endpoints when you need token budget control and content already formatted for language model injection.

## Known failure modes

- Missing required 'q' parameter returns a 400 validation error
- Token count outside 1024–32768 range may be rejected or clamped
- URL count outside 1–50 range may be rejected or clamped
- Invalid freshness value (not 'pd', 'pw', 'pm', 'py', or date range) returns an error
- Query returns no results — response body will be empty or minimal
- Upstream Brave index unavailability causes 5xx errors
- Payment failure or insufficient USDC balance prevents the call

## How this service works

Pre-extracted web content optimized for AI agents, LLM grounding, and RAG pipelines.

## Output

A distilled, token-bounded text corpus derived from up to 50 Brave Search result pages, trimmed to fit within the specified token budget (1024–32768 tokens), ready to inject directly into an LLM prompt as grounding context.

## Example request

```json
{
 "q": "artificial intelligence",
 "freshness": "pw",
 "maximum_number_of_urls": 10,
 "maximum_number_of_tokens": 8192
}
```

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "q"
 ],
 "properties": {
  "q": {
   "type": "string",
   "description": "Search query"
  },
  "freshness": {
   "type": "string",
   "description": "\"pd\", \"pw\", \"pm\", \"py\", or date range"
  },
  "maximum_number_of_urls": {
   "type": "number",
   "description": "Max URLs (1-50, default 20)"
  },
  "maximum_number_of_tokens": {
   "type": "number",
   "description": "Max tokens (1024-32768, default 8192)"
  }
 },
 "additionalProperties": true
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/brave-search-llm-context-d4d573af/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from brave.mpp.paywithlocus.com](https://www.zero.xyz/host/brave.mpp.paywithlocus.com/llms.txt)