← ALL TAGS

Tag: llm

5 articles tagged “llm” on Oak Oliver Engineering, newest first. The latest is “Getting Started with pageindex: Tree Indexes for Reasoning-Based RAG”.

Runtime

Getting Started with pageindex: Tree Indexes for Reasoning-Based RAG

A hands-on introduction to pageindex, the TypeScript port of PageIndex: turn a Markdown document into a tree of sections from the CLI and from code, thin the tree, add summaries through any OpenAI-compatible endpoint, and write the tree-search loop that answers questions from the right section. Offline first, with clear notes on which steps need an LLM.

October 3, 2026 10 min read
Read Story →
Runtime

Getting Started with kb: A Knowledge Base an LLM Writes for You

@oakoliver/kb turns notes, papers and web pages into a linked, Obsidian-compatible wiki: you ingest sources, an LLM compiles them into articles, and you search, lint and browse the result from the terminal. This guide goes from an empty folder to KB Studio, and shows which steps need an API key.

October 3, 2026 7 min read
Read Story →
Runtime

Stop Pasting Into ChatGPT. Build a Knowledge Base Your AI Can Actually Search.

I built a CLI tool that turns URLs, PDFs, and markdown into a searchable knowledge base with AI-powered Q&A. No vector database. No embedding service. Just files, BM25 search, and LLM synthesis — all from your terminal.

April 3, 2026 7 min read
Read Story →
Runtime

How kb Works: 27 Modules, One External Dependency, and a Compiler That Turns Raw Sources Into a Wiki

A deep dive into the architecture of kb — the CLI knowledge base tool. Hash-based incremental compilation, BM25 search, TTY-aware dual output, Zod-validated frontmatter, and a dependency graph that recompiles only what changed. 154 tests, 27 TypeScript modules, the full @oakoliver stack.

April 3, 2026 8 min read
Read Story →
Security

Server-Side API Key Injection: An LLM Proxy That Hides Your Keys and Meters Cost Per Request

How Vibe's AI proxy injects its Gemini key on the server so browser code never sees them, and why it meters each request's cost at the proxy, before and after the call, instead of working it out from logs later.

February 14, 2026 16 min read
Read Story →