SkillFoxx

Brave Search API Skills

Official Brave skills that teach an agent to call the Search API for web, news, images, videos, places and page extracts for LLMs

  • 196 ★ on the repository
  • Last commit
  • MIT
SkillOfficial

Extended access

Why this level

  • The skills direct the agent to send keyed requests to the Brave Search API, so query text goes to a third-party service
  • The key in an environment variable is visible to the agent and the commands it runs
All reasons and checks

brave/brave-search-skillsAuthor:Brave

Install

In your terminal, with SkillFoxx CLI

npx skillfoxx add skills/brave-search-api-skills

Detects the agents on your machine, shows the access level and pins the version.

Other ways to install

Access and checks

How we review

Extended accessThe tool runs code or makes network calls and only reads external data.

  • The skills direct the agent to send keyed requests to the Brave Search API, so query text goes to a third-party service
  • The key in an environment variable is visible to the agent and the commands it runs

This is third-party code. Review the repository files before installing.

What it does

Brave's repository holds 12 skills, one per Brave Search API endpoint, and the agent builds requests from their instructions using the key in `BRAVE_SEARCH_API_KEY`. `web-search` returns ranked results with snippets, `news-search` returns news with a freshness filter, `images-search` and `videos-search` find images and video, and `suggest` and `spellcheck` complete and correct a query. Three skills handle places: `local-place-search`, `local-pois` and `local-descriptions`. For RAG and grounding there are `llm-context`, which returns pre-extracted page content, and `answers`, compatible with the OpenAI SDK. The `bx` skill wraps all endpoints in one CLI with token-budgeted output. The guide also covers Goggles, custom ranking rules for results. The skills live in `skills/` and any agent that follows the Agent Skills standard can read them. Claude Code has a marketplace plugin with auto-updates, and for other agents the folders are placed with `curl` or copied.

Who it is for. Developers and agents that need web search through the Brave Search API with fresh news, images and place data.

Good fit when

  • An agent needs fresh results and page excerpts for an answer and the agent's built-in search does not fit
  • A RAG pipeline should feed the model pre-extracted page content through `llm-context`
  • An app needs query autocomplete and spelling correction: the fast `suggest` and `spellcheck`
  • Results must follow your own ranking rules, such as downranking or discarding domains with Goggles

Not a fit when

  • The agent works only with local project files and has no use for web search
  • A Brave key cannot be obtained: every skill calls the paid API, and `web-search` is included in the Search plan
  • You need a ready MCP server: Brave has a separate repository for that, listed in the catalog

Example request

Use Brave to find this week's news on a Claude release and return five links with short excerpts

Limitations

  • A Brave Search API key (https://api.search.brave.com) in `BRAVE_SEARCH_API_KEY` is required; access to endpoints depends on the plan.
  • The skills do not search themselves: they give the agent instructions and `curl` examples, so network access and permission to run commands are needed.
  • Agents can read environment variables, and the author advises storing the key in the agent's own config without writing it to the shell profile.
  • Cursor must be restarted from a terminal after exporting the key; reloading the window is not enough.

How to disable. In Claude Code run `/plugin uninstall brave-search-skills`. For other agents delete the skill folders from the skills directory and the `BRAVE_SEARCH_API_KEY` from the settings.

FAQ

How do I install the Brave skills in Claude Code?

Run `/plugin marketplace add brave/brave-search-skills`, then `/plugin install brave-search-skills@brave-search`. Without the marketplace, download the `skills/` folder with `curl` into `~/.claude/skills` or copy it after `git clone`.

Where should I store the Brave Search API key?

Preferably in the agent's config: for Claude Code it is `env` in `~/.claude/settings.json`, for Codex `shell_environment_policy` in `~/.codex/config.toml`. For Cursor `direnv` with an `.envrc` file works.

Which skill should I pick for ordinary search?

For AI tasks the author recommends `llm-context`, which returns already extracted content, and `web-search` suits a classic results list with links. For shell use with a token budget there is `bx`.

What are Goggles in Brave Search?

Custom ranking rules that raise or lower domains and patterns in the results. The `goggles` parameter goes into the search request.

Do the Brave skills work with Codex?

Yes. Install them with Codex's built-in skill installer from `https://github.com/brave/brave-search-skills` or copy them into `~/.agents/skills`. Set the key in `config.toml` or the shell profile.

Discussion

No questions or reviews yet.

Skills

Three search skills: four-source search scored by intent, deep thread tracing and clean Markdown extraction from pages.

SkillExtended access438

Two small skills: web search through your own browser with no paid API, and a preflight risk check before a meaningful change.

SkillExtended access81
Official

A web search skill with no MCP server: four command-line clients search the web and 16 vertical domains, in batches and with page extraction

SkillExtended access6.9K

Portable skill for the Crawl4AI CLI and SDK: JavaScript-heavy pages, structured extraction by CSS schemas and batch crawling.

SkillBroad access50

A web research skill: SearXNG searches, Camofox reads pages and CloakBrowser opens sites that check their visitors

SkillBroad access546

Three skills and four utilities that let an agent search the web, read pages, pull video subtitles and process media

SkillExtended access228

README in short

The README starts with getting a key and safe ways to set it in Claude Code, Cursor, Codex and OpenClaw. Installation per agent follows, then a table of all skills with endpoints, a `curl` quick start, a description of Goggles and documentation links.

README badge

Original SKILL.md

---
name: web-search
description: USE FOR web search. Returns ranked results with snippets, URLs, thumbnails. Supports freshness filters, SafeSearch, Goggles for custom ranking, pagination. Primary search endpoint.
---

# Web Search

> **Requires API Key**: Get one at https://api.search.brave.com
>
> **Plan**: Included in the **Search** plan. See https://api-dashboard.search.brave.com/app/subscriptions/subscribe

## Quick Start (cURL)

### Basic Search
```bash
curl -s "https://api.search.brave.com/res/v1/web/search?q=python+web+frameworks" \
  -H "Accept: application/json" \
  -H "X-Subscription-Token: ${BRAVE_SEARCH_API_KEY}"
```

### With Parameters
```bash
curl -s "https://api.search.brave.com/res/v1/web/search" \
  -H "Accept: application/json" \
  -H "X-Subscription-Token: ${BRAVE_SEARCH_API_KEY}" \
  -G \
  --data-urlencode "q=rust programming tutorials" \
  --data-urlencode "country=US" \
  --data-urlencode "search_lang=en" \
  --data-urlencode "count=10" \
  --data-urlencode "safesearch=moderate" \
  --data-urlencode "freshness=pm"
```

## Endpoint

```http
GET https://api.search.brave.com/res/v1/web/search
POST https://api.search.brave.com/res/v1/web/search
```

**Note**: Both GET and POST methods are supported. POST is useful for long queries or complex Goggles.

**Authentication**: `X-Subscription-Token: <API_KEY>` header

Report a problem