# Zen4-bit/Proxima

Multi-AI MCP Server - Connect ChatGPT, Claude, Gemini & Perplexity to your coding tools without any API

Repository: https://github.com/Zen4-bit/Proxima
Canonical: https://ross.abutalabs.com/products/proxima
Homepage: https://www.proximamcp.in/
Language: JavaScript
License: NOASSERTION
License Family: other
Last push: 2026-08-17T17:16:06+00:00

## Health v2 (maintenance only)
Score: 75/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 98, release rhythm 80, longevity 15
- inputs: {"age_days": 212, "days_push": 16, "days_rel": 56, "gap_med": 61, "n_releases_24m": 4}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1169, forks 163 (observed 2026-08-28T04:03:51.005402+00:00)

## What it is
Proxima is a local-first Electron application that runs a multi-AI Model Context Protocol (MCP) server, routing ChatGPT, Claude, Gemini, and Perplexity (via free browser sessions, BYOK, or local models like Ollama) into IDEs such as Cursor, VS Code, and Windsurf. It also ships an autonomous Python coding agent with self-healing error loops, Chrome DevTools automation, and SQLite-backed cross-session memory.

## Use cases
- use chatgpt claude and gemini inside cursor without paying for api keys
- connect multiple ai models to my ide via mcp for free
- run a local mcp server with browser login sessions instead of api subscriptions
- self-host an autonomous coding agent with self-healing error correction
- route ollama and lm studio local models into vs code
- give my ai coding assistant web search and multi-model consensus tools
- avoid api costs by reusing my existing chatgpt plus and claude subscriptions

## When to choose
- you already pay for AI chat subscriptions and want to reuse them in your IDE without API fees
- you want a privacy-first, fully local setup with no cloud relay or telemetry
- you need multi-model routing, consensus queries, and 40+ MCP developer tools in one server
- you want an autonomous agent with self-healing loops and persistent local memory

## When to avoid
- you need a permissively licensed dependency for commercial products - the license is non-commercial
- you require guaranteed-stable production API integrations rather than browser-session automation that may break with UI changes
- you need a headless server-side deployment - it is an Electron desktop application
- you only need a simple single-model API client

## Facets
- artifact type: application
- maturity: active
- function: mcp, agent-framework, llm-inference, developer-tools, chatbot
- domain: large-language-models, developer-tools, self-hosted
- platform: windows
- tags: mcp-server, browser-session-routing, byok, ollama, local-first, coding-agent, cursor, vscode, windsurf, sqlite-memory, ai-agents, desktop, macos, linux, electron

## Member repositories
- Zen4-bit/Proxima (main) score 75

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:51.005402+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:28:40.434572+00:00, confidence not recorded.
  - readme: https://github.com/Zen4-bit/Proxima (fetched 2026-08-28T04:03:51.005402+00:00, sha 49f8a02de981)
  - homepage: https://www.proximamcp.in/ (fetched 2026-08-29T12:34:35.584071+00:00, sha e73ac10aeedb)
  - site_page: https://www.proximamcp.in/documentation (fetched 2026-08-29T12:34:35.586978+00:00, sha de0e30475697)
  - site_page: https://www.proximamcp.in/about (fetched 2026-08-29T12:34:35.589206+00:00, sha 6796309983d1)
- Data as of 2026-08-30T08:39:29.467469+00:00.