mirror of
https://github.com/bendtherules/ask262.git
synced 2026-08-18 21:31:46 +00:00
70e42eb727ec9d0083a28f37cde9feba6ba769c3
- Consolidate all structural element breakdowns into a single `breakDownSection` function using an `alwaysBreak` flag. - Introduce `BREAKDOWN_TAGS` configuration array containing metadata (tag, alwaysBreak, title/id selectors) for each element type. - Replace previous multi‑phase approach with a unified sequential breakdown flow: `emu-clause` (always extracts children) → `emu-table` → `emu-grammar` → `td` → `p`. - Add inline markers `[Subsection available: title "X" at sectionid: ID]` where content is removed, enabling parent awareness. - Update documentation to reflect the new unified breakdown logic, tag table, and hierarchical ID format. - Adjust threshold handling, recursion depth, and metadata tracking to work with the new unified approach.
RAG Pipeline for Language Specification Exploration
This project implements a RAG-based AI chat agent to explore the ECMAScript specification and its implementation in engine262.
Prerequisites
- Node.js: Version 18+
- Ollama: Installed locally with an embedding model (e.g.,
nomic-embed-text) - OpenAI-compatible Endpoint: A hosted or local LLM service
Setup
-
Install dependencies:
npm install -
Prepare environment:
export OPENAI_API_BASE="your_endpoint_base_url" export OPENAI_API_KEY="your_api_key" -
Clone specification: (Ensure
./spec-built/multipagecontains the HTML files) -
Ingest data:
node ingest.mjsNote: This will take significant time as it generates local embeddings via Ollama for both the spec and the implementation.
-
Build graph:
node build_graph.mjs
Usage
Ask the agent questions about how code relates to the specification:
node agent.mjs "Explain how the 'if' statement works and show its implementation."
The agent will use tools to search the specification, explore the implementation code, and navigate the relationships between them using the graph.
Languages
HTML
49%
JavaScript
32.7%
TypeScript
18.3%