brave-search

Wrap Brave Search web and summarizer endpoints into structured records and summaries.

2|Updated Jan 4, 2026
One-click install
npx skills add https://github.com/akaihola/skills-akaihola --skill brave-search-akaihola
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: brave-search
Source: https://github.com/akaihola/skills-akaihola/tree/main/brave-search
Command: npx skills add https://github.com/akaihola/skills-akaihola --skill brave-search-akaihola

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires httpx>=0.27.0, and includes scripts (resource) and references (resource) components.

What problem does it solve?

Brave Search API users need a deterministic wrapper around Brave's web search and summarizer endpoints to produce structured results and concise narratives, without reimplementing the request logic or polling behavior.

Core Features & Use Cases

  • Web search results delivered as structured records (web, faq, discussions, news, videos) suitable for downstream reasoning and tooling.
  • Summarizer workflow that consumes a web-derived summarizer key and returns a flattened, coherent summary with optional inline references and enriched metadata.
  • End-to-end CLI and programmatic interfaces that enable deterministic execution, parameter normalization, and robust error handling.

Quick Start

Run a web search or summarizer workflow by invoking the CLI with uv run scripts/brave_search.py web --params-json '<JSON>' or uv run scripts/brave_search.py summarizer --params-json '<JSON>'.

Frequently Asked Questions about brave-search

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I get structured web search results from the Brave Search API?

Structured web search results from the Brave Search API are obtained by running the web workflow via CLI, which normalizes parameters and returns categorized records like news, videos, and discussions for downstream use.

How does the Brave Search summarizer workflow process queries?

The Brave Search summarizer workflow operates via a two-step sequence: it first obtains a summarizer key from web search results, then calls the summarizer endpoint and polls until completion to return a coherent summary.

Can I use httpx to execute deterministic Brave Search queries programmatically?

Yes, you can execute deterministic Brave Search queries programmatically using the provided CLI or programmatic interfaces, which coordinate parameter normalization and robust error handling over the httpx dependency.

What is the best way to handle Goggles parameters in Brave Search API requests?

Handling Goggles parameters in Brave Search API requests is managed automatically by the skill's deterministic wrappers, which normalize complex API surfaces and Goggles configurations into structured, consumable data.

Does the Brave Search summarizer return inline references with its synthesized summaries?

Yes, the Brave Search summarizer returns a flattened, coherent summary that includes optional inline references and enriched metadata alongside the synthesized narrative output.

Why do Brave Search summarizer calls require polling for completion?

Brave Search summarizer calls require polling for completion because the summarizer workflow is asynchronous; the deterministic wrapper coordinates this polling behavior after retrieving the initial summarizer key from web results.