# 1jehuang/jcode

The most RAM efficient harness

Repository: https://github.com/1jehuang/jcode
Canonical: https://ross.abutalabs.com/products/jcode
Homepage: https://jcode.sh
Language: Rust
License: MIT
License Family: permissive
Topics: ai, claude, cli, coding-agent, llm, mcp, openai, rust, terminal, tui, ai-agent, ai-coding-agent
Last push: 2026-08-26T22:36:20+00:00

## Health v2 (maintenance only)
Score: 78/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 99, release rhythm 87, longevity 17
- inputs: {"age_days": 240, "days_push": 7, "days_rel": 7, "gap_med": 0.0, "n_releases_24m": 165}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 18635, forks 2108 (observed 2026-08-28T04:11:26.908939+00:00)

## What it is
jcode is an open-source AI coding agent for the terminal, written in Rust, that lets developers run many agent sessions in parallel with very low RAM usage. It supports multiple model providers (Claude, OpenAI, Gemini, Copilot, Azure, local models) via OAuth or API keys, and offers a TUI, non-interactive runs, and a background server mode.

## Use cases
- run an AI coding agent in the terminal
- run dozens of coding agents in parallel cheaply
- use Claude or ChatGPT subscription with a CLI coding agent
- find a low-memory alternative to Claude Code
- delegate coding tasks to background AI agents
- run a coding agent with local models via Ollama

## When to choose
- you want a resource-efficient, open-source terminal coding agent
- you need to run many agent sessions in parallel on one machine
- you want to bring your own provider subscriptions or local models
- you value an inspectable, MIT-licensed agent harness

## When to avoid
- you need a GUI or IDE-integrated coding assistant
- you rely on a vendor's official support and hosted-only experience
- you don't want any agent running commands on your machine

## Facets
- artifact type: cli-tool
- maturity: active
- function: agent-framework, llm-inference, cli, terminal-ui, mcp, developer-tools
- domain: developer-tools, large-language-models
- platform: windows, cli, rust
- tags: coding-agent, ai-coding, tui, parallel-sessions, resource-efficiency, byo-model-provider, ai-agents, command-line, linux, macos

## Member repositories
- 1jehuang/jcode (main) score 78

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:26.908939+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 2026-08-29T17:01:21.069501+00:00, confidence not recorded.
  - readme: https://github.com/1jehuang/jcode (fetched 2026-08-28T04:11:26.908939+00:00, sha a4d557e73c0c)
  - homepage: https://jcode.sh (fetched 2026-08-29T07:59:28.944606+00:00, sha 36ca5909d267)
  - site_page: https://jcode.sh/docs (fetched 2026-08-29T07:59:28.954607+00:00, sha 3744a9e6b366)
  - site_page: https://jcode.sh/about (fetched 2026-08-29T07:59:28.958552+00:00, sha 22ebbd53d41f)
  - site_page: https://jcode.sh/pricing (fetched 2026-08-29T07:59:28.956868+00:00, sha 9774d731cf23)
- Data as of 2026-08-30T08:39:29.467469+00:00.