# zhnt/loushang

AI-native agent harness for coding workflows by python: multi-model LLM orchestration, stateful sessions, tool governance,   traceable delivery, and provider routing for GPT, Claude, DeepSeek, Qwen, Kimi, GLM, and MiniMax.

Repository: https://github.com/zhnt/loushang
Canonical: https://ross.abutalabs.com/products/loushang
Language: Python
License: Apache-2.0
License Family: permissive
Topics: agent, claude, claude-code, codex, coding, deepseek, glm, harness, kimi, minimax, python, skills, workflow, agentic, ai, chatgpt, vibe-coding, agent-harness, dynamic-workflows, parallel-agents
Last push: 2026-08-26T14:28:29+00:00

## Health v2 (maintenance only)
Score: 58/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 99, release rhythm 35, longevity 6
- inputs: {"age_days": 96, "days_push": 7, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1257, forks 205 (observed 2026-08-28T04:04:08.933306+00:00)

## What it is
Loushang is an AI-native agent harness for coding workflows, delivered as a Python CLI and terminal workbench (`loushang code`) with multi-model provider routing, persistent resumable sessions, and governed tool execution. It treats methods, sessions, tools, and extensions as runtime objects to make complex AI-driven work reliable, recoverable, and auditable.

## Use cases
- run an AI coding agent from the terminal
- route coding tasks across GPT, Claude, DeepSeek, Qwen, Kimi, GLM, and MiniMax models
- resume and fork long coding agent sessions
- govern which tools an AI agent can execute
- add project-level custom tools and hooks to an agent workflow
- trace and audit AI-assisted delivery of code changes

## When to choose
- you want a Claude Code-style terminal coding agent with multi-provider model routing
- you need persistent, resumable, auditable agent sessions for complex coding work
- you want to extend an agent harness with custom Python tools and hooks
- you prefer running an open-source coding agent locally from source

## When to avoid
- you need a production-stable tool - the project is in early development
- you want a GUI or IDE plugin rather than a CLI/terminal workbench
- you only need a simple single-model chatbot without session or tool governance
- you cannot install Python tooling or run projects from source

## Facets
- artifact type: cli-tool
- maturity: experimental
- function: agent-framework, llm-inference, cli, developer-tools, workflow-automation, chatbot
- domain: large-language-models, developer-tools
- platform: python, cli, cross-platform, windows
- tags: agent-harness, coding-agent, model-routing, multi-provider, persistent-sessions, tool-governance, terminal-ui, vibe-coding, extensions, ai-agents, command-line, automation, linux, macos

## Member repositories
- zhnt/loushang (main) score 58

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:08.933306+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 2026-08-30T05:07:13.746542+00:00, confidence not recorded.
  - readme: https://github.com/zhnt/loushang (fetched 2026-08-28T04:04:08.933306+00:00, sha 56873f32e12f)
- Data as of 2026-08-30T08:39:29.467469+00:00.