Prism-Shadow/penguin-harness

Unified and Stable RSI Framework
Automatically improve model × agent harness

English | 简体中文

PenguinHarness is an open-source framework for agent self-evolution (recursive self-improvement, RSI). The model stays fixed; the harness evolves: prompts, Skills, tools and hooks. It runs on your computer or server as a desktop app, a Web App, a CLI and an SDK.

• Every RSI algorithm in one framework. Default, OPRO, APE, ACE and AWM ship as plugins and run on the same built-in Benchmarks, scoreboard and snapshots.
• One sentence builds an agent app. Describe what you need, and an agent writes the app, its page and its run instructions.
• Cheap to run. A deliberately small toolset, tuned for open models: on our data-analysis suite, the best accuracy at 1/70 of Claude Code's cost.

Pick an algorithm, pick a Benchmark, and an agent improves another agent round by round. Every candidate is a new version, snapshotted before it changes, and every score lands in the Evaluation Center.

Every new Project includes seven Benchmarks, run as Harbor tasks in Docker: PenguinHarness Benchmark Sec A to Sec E (from rag-bench-essential, DeepSWE, AutomationBench, Terminal-Bench-Science and Terminal-Bench) and GDPevo Train and Test. More are in penguin-harness-benchmark.

Demo: a decision agent is evaluated, revises its working method once, and is retested on the same model. Its score goes from 84 to 100, and the cost of every run is on screen.

Build a legal review agent with a simple workspace where users upload contracts, track the review, and download a report with risky clauses, explanations, and suggested revisions.

The result: a legal review app. Users upload contracts, follow each file's review, and download a report of risky clauses with explanations and suggested revisions. Five sample reviews cost $0.0423 in model calls.

Browser automation Agents drive your own Chrome through the PenguinHarness Browser extension, or the desktop app's built-in browser. They read pages as text and act through short scripts, so a step costs few tokens. Tutorial	Workbenches Every agent can have its own hot-reloaded workbench: ask for one and the agent writes the page and its server code. The workbench appears as a tab beside the agent's chat. Tutorial
Remote machines Add a machine the server reaches over ssh. Conversations in a Workspace on that machine run on it, in the same window. Tutorial	Agent API One unified API connects any agent to any external app: send input over HTTP and read the run back as a stream of AMSP events. Tutorial

Replies can also carry UI: choices, forms, steps, Mermaid diagrams and live widgets (A2UI).

Each harness on the model it is usually paired with, on the same tasks:

Agents can also write and optimize their own Skills (docs).

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.

Download the installer from penguin.ooo/download: macOS 11+ (dmg), Windows 10+ (exe) or Linux x64 (AppImage, deb). The app opens signed in and needs no terminal.

Then run penguin web. It serves the Web App at http://127.0.0.1:7364 and prints a first-login link until a password is set; models are configured on the Models page. The desktop app and the CLI share ~/.penguin/data.

docker run -d --name penguin -p 127.0.0.1:7364:7364 -v penguin-data:/data hiyouga/penguinharness:latest
docker logs penguin       # the first-login link, openable as it stands on this machine

The image serves on port 7364 with its data on the /data volume, for linux/amd64 and linux/arm64; latest follows main, and each release is also tagged with its version. The example listens on loopback only. For a network-facing deployment, Compose and upgrades, see the Docker quickstart.

mkdir penguin-install
tar -xzf penguin-linux-x64.tar.gz -C penguin-install
./penguin-install/install.sh

mkdir penguin-install
tar -xzf penguin-darwin-arm64.tar.gz -C penguin-install
./penguin-install/install.sh

Expand-Archive penguin-win32-x64.zip -DestinationPath penguin-install
cd penguin-install
.\install.cmd

PenguinHarness Browser lets agents drive the tabs you hand them in your own Chrome.

The same engine, scriptable, and made to be driven by agents:

penguin config model add --provider deepseek --model-id deepseek-flash --api-key sk-... --set-default
penguin run -m "Create hello.txt containing Hello, Penguin"   # one-shot task
penguin chat       # interactive REPL (/compact, /clear, /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);
  }
}

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

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

Working on this repository with a coding agent: its development Skills are in .agents/skills/ (penguin-harness-dev, penguin-harness-frontend, penguin-harness-manual-test). Claude Code loads them with no install, through the .claude → .agents symlink; for another agent, link or copy .agents/skills/* into the directory it reads Skills from.

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

Thanks to everyone who has contributed to PenguinHarness!

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}
}

This repo benefits from:

• MinGit: the POSIX shell and Git bundled on Windows
• GenericAgent: built-in browser automation

See THIRD-PARTY-NOTICES.md for their licenses.

Apache-2.0 © 2026 Prism Shadow

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