#ollama
Cloudflare Access Forward Proxy
nextgarch
Local HTTP proxy that adds CF-Access-Client-Id / -Secret headers to upstream requests. Lets VS Code chat extensions (which don't support custom headers) reach a Cloudflare-Access-gated OpenAI-compatible endpoint via http://127.0.0.1:.
Hawkeye – LLM Token Tracker
BlueJai
Track Claude Code and Codex CLI usage in real time. Session limits, weekly budgets, auto-switching, customizable status bar, and live API — all in your VS Code sidebar.
Synalux — Memory-Augmented AI
synalux-ai
Local + Cloud AI with 20 multimodal tools, multi-agent orchestration, Prism memory, and 12-language support. Works with Ollama (offline) or OpenRouter (cloud). HIPAA-compliant healthcare workflows.
Ollama Cloud Usage
Ollama-cloud-usage
View Ollama Cloud usage and limits in VS Code.
Reflexor: Local AI Coding Assistant
reflexor-app
Local-first AI coding assistant for VS Code and Cursor with codebase chat, RAG search, inline completion, and approval-first Agent workflows.
ctxkeeper — Never Lose Your AI Chat Context Again
ctxkeeper
Persistent memory + opt-in token-saving optimizer for GitHub Copilot. Local storage, zero telemetry, single universal vsix.
Navy AI Coder
Zrnge
AI coding assistant for VS Code — agentic tool-use, file edits with diff approval, multi-provider (Ollama, OpenAI, compatible APIs).
TinyFalcon
FalconRising
Your data, your compute. Local-first agentic coding assistant. Requires Ollama (free) running locally — works fully offline.
The Lean Prompt – Local Prompt Optimizer
zulkarnain-shah
Optimize prompts 100% on-device using local Ollama LLMs. Save tokens, reduce AI costs, stay private.
Heap Code
heapcode
Model-agnostic AI coding assistant. Works with any OpenAI-compatible API — Ollama, OpenAI, OpenRouter, Groq, LM Studio, vLLM, and more.
Lunaris Code
Lunaris
Your codebase, moving at the speed of thought.
OmniCode Proxy — Local AI Server
allwin-antony
Expose internal IDE language models (Gemini, Claude, etc.) as a local OpenAI-compatible HTTP API server. Like Ollama, but powered by the models already inside your IDE.