Enayat-Hassani

Yavar Extension — Git skill for Claude Code

Git community

A Chrome extension that turns free web chatbots (ChatGPT, Claude, Gemini) into a full workspace sidebar.

How to install Yavar Extension

This entry records only its repository, not the path inside it, so there is no exact command to give. Open Enayat-Hassani/Yavar-Extension and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Yavar Extension does

A Chrome extension that turns free web chatbots (ChatGPT, Claude, Gemini) into a full workspace sidebar. Features single-click text and screenshot routing, autonomous web research agents, GitHub codebase deep-dives, interactive architecture diagramming, and built-in answer history—no API keys required.

Alternatives in Git

Triage Issues — Triage open GitHub issues — split into bugs vs features, rank by severity/opportunity, and flag under-specifie 12.6k ★
Git MCP — GitHub Repo stars Free open-source remote MCP server for any GitHub project Git integration MCP 7.8k ★
Fullstack Dev Skills — by jeffallan - A comprehensive Claude Code plugin with 65 specialized skills covering full-stack development a 7k ★

README

Yavar - Your AI Sidekick

A Chrome extension that embeds ChatGPT, Claude, and Gemini in a sidebar — so you can select anything on any page and route it to the AI, capture screenshots, analyze GitHub repositories with a deep-dive agent, run a web research agent, and keep your AI answers in a searchable history.

What it does

**AI sidebar** — opens `gemini.google.com`, `chatgpt.com`, or `claude.ai` in a full-viewport sidebar, with a model switcher (and support for adding your own custom models). Send selected text, screenshots, or whole-page content straight to the model.

**Floating menu** — select text on any page and a small action menu appears:

Send — send the selection straight to your default AI platform.
Explain — send it wrapped in a "Guided Learning" prompt.

**GitHub analysis** — on any GitHub repository, generate a structured learning prompt with the file tree and README context. Unauthenticated by default; optionally add a GitHub token in Settings to lift the 60 req/hr rate limit.

**GitHub deep-dive agent** — pick a repository to scan, then ask questions about it. The agent reads the codebase and answers with file references, with a live working-status bar.

**Web research agent** — research any topic: the agent performs **SEARCH + READ** across the web and synthesizes an answer in the sidebar. Toggle **Deep research mode** in Settings for deeper coverage.

**Repo file browser** — browse any GitHub repository's file tree right in the sidebar, search/filter it, and jump to (or quick-add) the active file.

**Architecture diagrams** — generate a Mermaid **architecture diagram** of the current repository, rendered interactively in the sidebar.

**History & saved answers** — capture the AI's last answer and keep it in a saved-answers history panel you can return to, copy, or clear.

**Notes panel** — a built-in CodeMirror-powered scratchpad inside the sidebar, toggled with the notes shortcut.

**Auto-submit & auto-paste** — send selected tex