# Danau5tin/multi-agent-coding-system

Reached #13 on Stanford's Terminal Bench leaderboard. Orchestrator, explorer & coder agents working together with intelligent context sharing.

Repository: https://github.com/Danau5tin/multi-agent-coding-system
Canonical: https://ross.abutalabs.com/products/multi-agent-coding-system
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2025-11-03T12:22:00+00:00

## Health v2 (maintenance only)
Score: 33/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 50, release rhythm 14, longevity 26
- inputs: {"age_days": 368, "days_push": 303, "days_rel": 362, "gap_med": null, "n_releases_24m": 1}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1438, forks 187 (observed 2026-08-28T04:04:43.885002+00:00)

## What it is
A Python multi-agent AI coding system where an orchestrator agent dispatches explorer and coder subagents with shared knowledge artifacts, ranking #13 on Stanford's TerminalBench leaderboard. It also supports RL training of agent models across distributed Docker environments.

## Use cases
- automate coding tasks with multiple cooperating AI agents
- benchmark an AI agent on TerminalBench
- orchestrate explorer and coder subagents with shared context
- train a coding agent model with reinforcement learning
- delegate software implementation work to LLM agents
- run concurrent agent rollouts in Docker environments

## When to choose
- you want a proven multi-agent coding pipeline with strong benchmark results
- you need orchestrator/subagent delegation with reusable knowledge artifacts
- you want to RL-train an agent model at scale with open training code

## When to avoid
- you need a simple single-agent coding assistant with minimal setup
- you want a polished GUI product rather than a framework
- your project cannot use LLM API costs or GPU training infrastructure

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, llm-inference, llm-training, developer-tools
- domain: artificial-intelligence, large-language-models, developer-tools
- platform: python, cross-platform, cli
- tags: multi-agent, orchestration, terminal-bench, coding-agents, context-sharing, reinforcement-learning, ai-agents, automation, docker

## Member repositories
- Danau5tin/multi-agent-coding-system (main) score 33

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:43.885002+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 2026-08-30T04:36:38.383338+00:00, confidence not recorded.
  - readme: https://github.com/Danau5tin/multi-agent-coding-system (fetched 2026-08-28T04:04:43.885002+00:00, sha d2e15a97a7ec)
- Data as of 2026-08-30T08:39:29.467469+00:00.