robertpiosik/CodeWebChat

Blazing fast AI coding with static context

CodeWebChat – AI‑assisted coding inside your editor

What it does – CodeWebChat (CWC) is a VS Code extension (also available for Cursor and Antigravity) that lets you talk to LLMs about the code in your workspace. By checking files in the explorer you build a static context that is sent to the model, which then returns edits, completions, search results, commit messages, or transcriptions. The extension formats these interactions as markdown prompts so the model knows exactly what to output.

Key features

  • Privacy‑first – everything runs locally; no code or prompts leave your machine.
  • Automatic context selection – CWC can invoke an Agentic Search prompt that filters the workspace to only the files relevant to the current task.
  • Prompt caching – unchanged files are sent first, recent changes later, improving cache hits.
  • Prompt templates – a set of ready‑made markdown prompt types (Agentic Search, Editing, Asking, Intelligent Search, Code‑at‑Cursor, Commit Messages, Patch Repair, Voice Input) for web, API, and CLI use.
  • Rich UI commands – save/restore selections, copy markdown, generate commit messages, manage workspace checkpoints, and apply LLM responses directly.

How you use it

  1. Install the extension from the VS Code Marketplace or Open VSX.
  2. Tick the files you want the model to see.
  3. Choose a command (e.g., Edit, Ask, Code at Cursor). CWC builds the appropriate markdown prompt and sends it to the configured LLM.
  4. The model replies with structured markdown (file headings, code blocks, or XML tags). Press Apply Chat Response to write the changes back to your project.

Who it’s for – developers who want fast, local AI assistance for code generation, refactoring, documentation, or commit‑message drafting without relying on cloud services.

Build yourself – The repo contains a monorepo with a Node 22 + pnpm setup. After cloning, run pnpm install, build the editor app (cd apps/editor && pnpm run build), and install the produced .vsix file in VS Code.

All details are taken directly from the README; no additional claims are made.

Related

  • Project
    mufeedvh/code2prompt

    A context engineering tool that converts entire codebases into structured LLM prompts, automating the process of gathering and formatting code for AI models.

  • Project
    Houseofmvps/codesight

    A zero-dependency tool that pre-compiles codebase structure and documentation into compact context maps and wikis to drastically reduce token consumption for AI coding assistants.

  • Project
    kunal12203/GrapeRoot

    An open-source launcher for AI coding assistants that builds a local semantic graph of a codebase to pre-load relevant context into prompts, reducing token costs and response times.

  • Dispatch
    Maximizing Claude Code Session Value and Token Efficiency

    Optimizing Claude Code sessions requires strategic use of prompt caching, context management, and subagents to reduce token costs and prevent cache misses.

  • Project
    jfrog/boost

    Boost is a command wrapper that compresses noisy shell output into structured context for AI coding agents, reducing token costs without sacrificing task performance.