LlamaIndex ‘legal-kb’: Agentic Retrieval over Index v2 with retrieve, find, read, and grep Tools
LlamaIndex ships agentic retrieval tools. Agents can now grep, find, and read documents like a filesystem.
Why it matters
LlamaIndex released a reference implementation (legal-kb) demonstrating agent-native document retrieval at scale. This shows RAG evolution from vector search to multi-tool agent workflows—directly applicable to enterprise knowledge work.
The key facts
11 to knowLlamaIndex legal-kb: public reference app for agentic document retrieval
Tools exposed: retrieve (hybrid semantic search), find, read, grep
Built on Index v2 with automatic per-file versioning
Tech stack: TanStack Start, AI SDK 6 (ToolLoopAgent), Prisma, WorkOS
Visual citations and filesystem-style agent access to knowledge base
LlamaIndex released legal-kb reference app
Exposes retrieve (hybrid semantic search), find, read, and grep as agent tools
Automatic per-file versioning and visual citations built in
Stack: TanStack Start, AI SDK 6 (ToolLoopAgent), Prisma, WorkOS
Demonstrates agentic access patterns to document knowledge bases
Index v2 integration
Go to the source
MarkTechPostmarktechpost.com
Publisher excerpt: LlamaIndex’s legal-kb is a public reference app that gives agents filesystem-style access to a document knowledge base on Index v2. It exposes retrieve (hybrid semantic search), find, read, and grep as tools, with automatic per-file versioning and visual citations. The stack is TanStack Start, AI…