Llm Context

If you're building RAG pipelines or AI agents, this is your connector to Brave's search index. It skips the snippets and returns actual extracted web content: text chunks, tables, code blocks, all pre-ranked and ready to feed your LLM. The real power is in the Goggles feature, which lets you filter sources inline (official docs only, no paywalls, academic sources) so you control what grounds your model. You get granular knobs for token budgets and snippet counts, plus local search support when you need location-aware results. It's faster than their full answer engine and designed specifically for tool calls and agentic workflows. One oddity: the threshold modes (strict/balanced/lenient) affect relevance filtering, but you'll need to experiment to see what works for your use case.

Install to Claude Code

npx -y skills add brave/brave-search-skills --skill llm-context --agent claude-code