PenguinHarness

Your automated agent builder lives on your desktop, connected to 1,000+ models.

English | 简体中文

Why PenguinHarness

With LangChain, you build agents by hand — at 1× speed.
With PenguinHarness, agents build agents — at 100×.

Three reasons, in deliberate order — from task quality, to how agents get built, to how they keep improving.

1. 🏆 Outstanding results at tens of times less cost

A deliberately minimal toolset over clean low-level interfaces: fewer tool calls, fewer tokens — deeply tuned for open models like DeepSeek. Each harness on the model it is normally paired with, same tasks, head-to-head:

Best accuracy on data analysis — at 1/70 of Claude Code's cost.

2. ⚡ One sentence, and an agent builds your agent app

Type one sentence, and an agent builds the complete agent application for you — scaffold, code, and run instructions, end to end:

Collect the docs from https://github.com/ericbuess/claude-code-docs and build a RAG app that answers Claude Code questions as a configuration expert, citing its sources.

And this is the finished product — a docs expert with retrieval, cited sources that link to the original files, and example questions built in:

https://github.com/user-attachments/assets/9b7033e8-f08a-4c3f-bd33-547896664e6e

And generating this entire RAG app burned just $0.02 (¥0.2) of tokens — on DeepSeek V4 Pro.

3. 🧬 Self-evolution: it gets stronger with use

With PenguinHarness Skills, an agent evaluates and optimizes itself: run the benchmark, find the lost points, ship version N+1 — with a snapshot before every round, and every request observable in the Trace view.

https://github.com/user-attachments/assets/922d13a6-5ffc-4685-9a39-352f02f9afc0

Built-in Skills

Four Skill groups ship in the box (docs); agents can also write and optimize their own:

Group	Skills
Office Productivity	data-analysis, firecrawl
Software Development	web-design, software-engineering
AI App Development	penguin-sdk, penguin-cli, agenthub-models, vllm, ollama, llamafactory
Agent Tuning	agent-creation, benchmark-design, agent-evaluation, agent-optimization

Supported Models

Model	Providers
DeepSeek V4	DeepSeek, OpenRouter, Fireworks AI, SiliconFlow, Qwen Token Plan
Kimi K3	Moonshot AI, OpenRouter, Qwen Pay-As-You-Go
GLM 5.2	Z.AI, OpenRouter, Fireworks AI, SiliconFlow, Qwen Token Plan, Qwen Pay-As-You-Go
Hunyuan 3	OpenRouter
Qwen 3.8 Max	Qwen Token Plan (preview)
GPT 5.6	OpenRouter
Gemini 3.6 Flash	Google Gemini, OpenRouter
Claude 5	Anthropic, OpenRouter

Each family's latest generation only — the app's Models page lists every built-in preset, and any OpenAI-protocol endpoint works too: pick a preset, or point a custom endpoint at any of the 1000+ online and local models.

Requirements

Requirement	Supported
OS	Linux, macOS, Windows 10+
Architecture	x64, arm64
Runtime	bundled by the one-line installer (npm installs need Node >= 24)
Model	an API key for at least one model

Installation

🌐 Web App — for humans

🚀 Install and launch the full experience (multi-session chat, agent/skill/model management, usage stats, Trace observability, evaluation center):

curl -fsSL https://penguin.ooo/install.sh | sh
penguin web        # start the service and open http://127.0.0.1:7364 (first login: admin / penguin-2026)

🪟 On Windows (PowerShell): irm https://penguin.ooo/install.ps1 | iex

📦 Or via npm: npm install -g @prismshadow/penguin-cli. Configure models on the in-app Models page, then chat.

🤖 CLI & SDK — for agents

The same engine, scriptable — made to be driven by agents (and agents building agents):

penguin config model add --provider deepseek --model-id deepseek-v4-pro --api-key sk-... --set-default
penguin run -m "Create hello.txt containing Hello, Penguin"   # one-shot task
penguin chat       # interactive REPL (/compact, /exit, Ctrl-C to interrupt)
penguin server     # headless service (same API the Web App uses)

import { createAgent, isCompleteModelMessage, userText } from "@prismshadow/penguin-core";

const agent = await createAgent({ agentId: "default_agent" });
const session = await agent.createSession({ workspaceDir: process.cwd() });

for await (const output of session.run([userText("Create hello.txt containing hi")], {
  approve: async () => "allow", // per-tool-call approval
})) {
  if (isCompleteModelMessage(output) && output.payload.type === "text") {
    console.log(output.payload.text);
  }
}

Roadmap

• Public release of the benchmark suite
• Desktop app
• Windows support
• Agent company and templates
• Company-level self evolving
• OpenShell integration (permission-governed shell)
• More to come…

Development

pnpm install && pnpm build   # build first: core's exports point at dist/
pnpm dev                     # backend + web app together (prefixed logs, deps built once)

See CONTRIBUTING.md for the full workspace guide: dev commands, quality gates, repo layout, and the changelog rule.

Citation

If you use PenguinHarness in your research, please cite:

@software{penguinharness2026,
  author  = {{PrismShadow Team}},
  title   = {PenguinHarness: Efficient Self-Improving Harness for Everyone},
  year    = {2026},
  url     = {https://github.com/Prism-Shadow/penguin-harness},
  license = {Apache-2.0}
}

License

Apache-2.0 © 2026 Prism Shadow

Built with ❤️ by Yaowei Zheng (author of LlamaFactory), the PrismShadow AI Team, and Fable 5.