docs: update AGENTS.md

replace LlamaIndex with LanceDB, update prerequisites, CLI commands, and expand project structure in AGENTS.md

AGENTS.md

@@ -5,7 +5,7 @@
 **Ask262** is a RAG-based AI chat agent for exploring the ECMAScript specification and its implementation in [engine262](https://github.com/bendtherules/engine262) (a JavaScript engine written in JavaScript).
 
 The agent combines:
-- **Vector search** (via LlamaIndex + Ollama embeddings) to find relevant spec sections
+- **Vector search** (via LanceDB + Ollama embeddings) to find relevant spec sections
 - **Knowledge graph** (via Graphology) mapping spec sections to implementation functions
 - **LLM reasoning** (via OpenAI-compatible API) to answer questions about JavaScript internals
 
@@ -13,73 +13,111 @@ Users ask questions like: *"How does the if statement work?"* or *"Which functio
 
 ## Prerequisites
 
-- **Node.js** 18+ with Bun runtime
+- **Node.js** 18+ with Bun runtime (packageManager: bun@1.2.8)
-- **Ollama** installed locally with `nomic-embed-text` model
+- **Ollama** installed locally with embedding model (e.g., `qwen3-embedding:0.6b`)
-- **NVIDIA API key** (or other OpenAI-compatible endpoint) in `config.json`
+- **OpenAI-compatible endpoint** (NVIDIA or other) configured in `config.json`
 
 ## Project Structure
 
 ```
 ask262/
 ├── agent.ts                  # Main ReAct agent - answers user queries
-├── agent-tools/          # Agent tool implementations
+├── agent-tools/              # Tool implementations (spec retriever, graph explorer)
-├── constants.ts          # Directory paths and file locations
+│   ├── index.ts              # Tool exports
+│   ├── specRetriever.ts      # Vector search tool
+│   ├── sectionRetriever.ts   # Section chunk retrieval tool
+│   ├── graphExplorer.ts      # Knowledge graph navigation tool
+│   └── reranker.ts           # Document reranking utility
+├── constants.ts              # Directory paths and model configs
 ├── config.json               # API keys and endpoints (user-created)
-├── setup/
+├── setup/                    # Data ingestion and graph building
 │   ├── ingest.ts             # Ingests spec HTML into vector index
-│   ├── buildGraph.ts     # Builds knowledge graph (spec → code mappings)
+│   ├── buildGraph.ts         # Builds knowledge graph
-│   └── stripSpecContainer.ts   # Cleans spec HTML files
+│   ├── stripSpecContainer.ts # Cleans spec HTML files
-├── spec-built/multipage/ # ECMAScript spec HTML files (ecmarkup output)
+│   ├── htmlAddInternalMethodLink.ts  # Adds internal method links
-├── engine262/src/        # JavaScript engine implementation code
+│   ├── text-splitters/         # Text chunking utilities
-├── storage/              # Vector index persistence (LlamaIndex)
+│   └── utils/                  # Formatting utilities
-└── graphology/           # Knowledge graph JSON file
+├── test/                     # Manual verification tests
+│   ├── manual/
+│   │   ├── verify-db.ts      # Verify database contents
+│   │   └── test-spec-retriever.ts
+├── spec-built/multipage/     # ECMAScript spec HTML files
+├── engine262/src/            # JavaScript engine implementation
+├── storage/                  # Vector index persistence (LanceDB)
+├── graphology/               # Knowledge graph JSON file
+├── biome.json                # Biome formatter config
+├── tsconfig.json             # TypeScript strict config
+└── package.json              # Bun-based dependencies
 ```
 
-**Key Files:**
-- `agent.ts`: ReAct agent with two tools - `spec_retriever` (vector search) and `graph_explorer` (graph navigation)
-- `setup/ingest.ts`: Parses HTML files, extracts `emu-clause` sections, chunks them, creates embeddings
-- `setup/buildGraph.ts`: Parses spec sections and code functions, creates nodes/edges showing which functions implement which spec sections
-- `constants.ts`: Defines `STORAGE_DIR`, `SPEC_DIR`, `CODE_DIR`, `GRAPH_FILE`
-
 ## Commands
 
-**Manual:**
 ```bash
-bun run setup/ingest.ts     # Direct ingest execution
+bun run ingest          # Ingest spec HTML into vector index
-bun run setup/buildGraph.ts  # Direct graph build
+bun run build           # Build knowledge graph (spec → code mappings)
-bun run agent.ts "Your question here"  # Direct agent execution
+bun run lint            # Check code with Biome
+bun run lint:fix        # Fix auto-fixable issues
+bun run format:fix      # Format code with Biome
+bun run type-check      # TypeScript check (no emit)
+bun test                # Run all tests
+bun run agent "Query"   # Run agent with question
 ```
 
-## File / Folder Naming
-Name all folders in kebab-case. Ex - hello-world
-Name all files in camelcase. Ex - helloWorld
-
 ## Code Style Guidelines
 
 ### Imports & Modules
 - Use ES modules (`import/export`), never CommonJS
 - Node.js built-ins: `import fs from "node:fs"` (with `node:` prefix)
 - Third-party: `import * as cheerio from "cheerio"`
-- Type: `"type": "module"` in package.json
+- Package type: `"type": "module"` in package.json
 
 ### Formatting (Biome)
 - **Indent**: 2 spaces (not tabs)
-- Excluded: `spec-built/`, `engine262/`, `graphology/` (external/vendor)
+- **Line width**: 80 characters
+- **Line ending**: LF
+- **Excluded directories**: `spec-built/`, `engine262/`, `graphology/` (external/vendor)
 
 ### TypeScript
 - **Target**: ES2022
-- **Strict mode**: Enabled
+- **Strict mode**: Enabled with `strict: true`
 - **Module resolution**: Node
 - Explicit types for function parameters and return values
 - Use `as const` for literal arrays
 - Avoid `any` - use proper types or `unknown`
+- JSON imports: `resolveJsonModule: true`
+
+### Naming Conventions
+- **Folders**: kebab-case (e.g., `agent-tools`, `text-splitters`)
+- **Files**: camelCase (e.g., `buildGraph.ts`, `specRetriever.ts`)
+- **Functions**: camelCase (e.g., `createGraphExplorerTool`)
+- **Constants**: UPPER_SNAKE_CASE or camelCase for exported constants
+- **Types/Interfaces**: PascalCase with descriptive names
 
 ### Comments & Documentation
 - JSDoc for public functions explaining purpose, params, return values
-- Inline comments for complex logic or non-obvious decisions
+- Use `/** */` for documentation blocks
-- Use `//` for implementation notes, `/** */` for documentation
+- Use `//` for inline implementation notes
+- Document complex logic or non-obvious decisions
 
-## Important notes
+### Error Handling
+- Use `try/catch` for async operations with meaningful error messages
+- Validate environment variables (e.g., `NVIDIA_API_KEY` in config.json)
+- Check file existence before reading
+- Log warnings for missing configuration rather than failing silently
 
-1. Always use Typescript for implementation. Don't use plain javascript.
+## Key Dependencies
-<!-- 2. Documentation for llamaIndex is  -->
+
+- **LangChain**: Agent framework and LLM integration (`langchain`, `@langchain/*`)
+- **LanceDB**: Vector storage for embeddings (`@lancedb/lancedb`)
+- **Ollama**: Local embeddings (`@langchain/ollama`)
+- **Graphology**: Knowledge graph library (`graphology`)
+- **Cheerio**: HTML parsing (`cheerio`)
+- **Biome**: Linting and formatting (`@biomejs/biome`)
+
+## Important Notes
+
+1. Always use TypeScript for implementation - no plain JavaScript
+2. Run type-check before committing: `bun run type-check`
+3. Ingest and build commands can take significant time due to local embedding generation
+4. The agent relies on `config.json` for API credentials (not committed to git)
+5. External directories (`spec-built/`, `engine262/`, `graphology/`) should not be modified by linting/formatting