How to use LLM Context Search in your AI agent

What you'll get

LLM Context delivers pre-extracted web content (text, tables, code) optimized for grounding AI responses. Instead of links and snippets, you get the actual page content your AI can reason over. Perfect for building AI agents, research tools, or any app that needs real-time web knowledge without the hassle of scraping.

Step 1 — Get the skill source URL

The LLM Context Search skill lives at:

Step 2 — Add it to your AI agent

Copy one of these prompts and paste it into your AI tool. The agent will fetch and install the skill for you.

Add the agent skill from https://github.com/brave/brave-search-skills/tree/main/skills/llm-context to my project.

Install this agent skill: https://github.com/brave/brave-search-skills/tree/main/skills/llm-context

Add this skill to .codex/skills/: https://github.com/brave/brave-search-skills/tree/main/skills/llm-context

Add the skill at https://github.com/brave/brave-search-skills/tree/main/skills/llm-context to my .gemini/skills/ directory.

Install this agent skill from https://github.com/brave/brave-search-skills/tree/main/skills/llm-context

Add this skill: https://github.com/brave/brave-search-skills/tree/main/skills/llm-context

Step 3 — Try these prompts

Once the skill is installed, ask your AI things like:

• Building an AI assistant that needs to answer questions with current web data
• You need your AI to cite real sources without you manually finding and copying content
• Researching a topic and want structured, AI-ready content instead of browsing links
• Limit AI answers to specific trusted sources (official docs, academic papers, etc.)

What this skill does

• Extracts clean text, tables, and code blocks from search results automatically
• Returns pre-processed web content optimized for AI assistants to read
• Filters results by relevance and lets you control sources with custom ranking rules
• Supports location-aware search for local business queries
• Adjusts content size based on your needs (simple facts vs. deep research)