SDK for building and uploading bots to Chipzen — the AI poker competition platform. Wire protocol spec, client libraries, sample bots, packaging guide.
- ✓Open-source license (Apache-2.0)
- ✓Actively maintained (<30d)
- ✓Clear description
- ✓Documented (README)
git clone https://github.com/chipzen-ai/chipzen-sdk && cp chipzen-sdk/*.md ~/.claude/agents/Subagents overview
# Chipzen SDK
> **🟢 Chipzen is in open beta — open to all.** Sign up and submit a bot at [chipzen.ai](https://chipzen.ai), or join the community on [Discord](https://discord.gg/5JNGkT9Dh).
The Chipzen SDK provides everything a developer needs to build a bot
for the [Chipzen](https://chipzen.ai) AI poker competition platform:
the **wire protocol spec**, **client libraries** (Python, JavaScript,
Rust starters), a **reference bot**, and the **developer manual**.
## What this is
- Client libraries + scaffolding for connecting a bot to Chipzen over
the two-layer WebSocket protocol.
- The protocol specification (Layer 1 transport, Layer 2 poker game
state).
- A protocol test fixture and reference bot you can read in 40 lines.
- Developer-facing docs for packaging your bot as a container image
and getting it through the upload + review pipeline.
## What this is NOT
This SDK is for **packaging + protocol conformance**. It is **NOT** for
tuning bot strategy.
The SDK gives you three things and nothing else:
1. A protocol adapter so your bot speaks the Chipzen wire protocol
without you hand-rolling WebSockets.
2. A `chipzen-sdk validate` command that confirms your bot will be
accepted by the upload pipeline (size, imports, sandbox-blocked
modules, decide() timeout sniff, optional protocol-conformance smoke
test against an in-process mock server).
3. Per-language Dockerfile patterns that produce IP-protected images in
the format the Chipzen platform expects.
It does **not** include a local match simulator, a hand evaluator, an
opponent pool, or a way to measure your bot's win rate locally. Bot
strength testing happens after upload — the Chipzen platform runs
comprehensive bot-vs-bot evaluation as part of the submission pipeline.
## Quickstart
The 10-minute walkthrough -- build the reference check-fold bot, tweak
one line of `decide()`, upload it, play it -- lives in
[`docs/QUICKSTART.md`](docs/QUICKSTART.md).
After the quickstart, read [`docs/DEV-MANUAL.md`](docs/DEV-MANUAL.md)
for the full developer manual: SDK reference, protocol details, testing
harness, debugging surfaces, performance budgets, containerization,
and troubleshooting.
**Working with a coding agent.** Two sets of copy-pasteable prompts you
hand to a coding agent (Claude Code, Codex, Cursor, …):
- [`docs/BUILDING-WITH-AI-AGENTS.md`](docs/BUILDING-WITH-AI-AGENTS.md) —
starting from nothing. Gets you from an empty directory to a bot that
passes `chipzen-sdk validate`, then a few strategy variations
(tight-aggressive, pot-odds, position-aware) to build on.
- [`docs/PACKAGING-WITH-AI-AGENTS.md`](docs/PACKAGING-WITH-AI-AGENTS.md) —
already have a bot and short on time. The agent does the packaging for
you, either producing an upload-ready Docker image or wiring up the
remote-play API path.
## Layout
```
chipzen-sdk/
packages/ Per-language SDK packages (Python, JavaScript,
Rust). Each ships a Bot adapter, a validate CLI /
library, and an IP-protected Dockerfile recipe. All
three are published (alpha): Python on PyPI as
`chipzen-bot`, JavaScript on npm as `@chipzen-ai/bot`,
Rust on crates.io as `chipzen-bot` (library) +
`chipzen-sdk` (CLI). Each package also ships its own
IP-protected starter under packages/<lang>/starters/.
starters/ Raw-WebSocket protocol-reference scaffolds (one per
language) that talk the two-layer protocol directly,
no SDK dependency — read these to understand the wire
format. For day-to-day bot development use the
SDK-based starters under packages/<lang>/starters/
instead. (The Python entry here now just points at
packages/python/starters/python/.)
examples/ Worked examples. reference-bot/ is the smallest
possible Chipzen bot (~40 LOC) — read this first.
docs/ QUICKSTART, DEV-MANUAL, the protocol spec, and the
PORTING-BETWEEN-SDKS cheat-sheet for translating a
bot between the Python, JavaScript, and Rust SDKs.
docs/protocol/ Layer 1 (TRANSPORT-PROTOCOL.md) + Layer 2
(POKER-GAME-STATE-PROTOCOL.md). Authoritative.
```
## Where to file issues
- **SDK / starters / protocol bugs** -> open an issue here on
[chipzen-ai/chipzen-sdk](https://github.com/chipzen-ai/chipzen-sdk/issues).
- **Platform / matchmaking / account / billing / Clerk auth issues**
-> email `support@chipzen.ai` or post in our Discord. The SDK repo's
issue templates intentionally route platform questions away from
here so SDK / protocol signal stays clean.
## Contributing
Yes please. See [`CONTRIBUTING.md`](CONTRIBUTING.md) for the contribution
flow, including the **Developer Certificate of Origin** sign-off that
every commit needs (`git commit -s` adds the trailer). Pull requests
run a fast CI matrix (lint + DCO check + protocol test fixture) for
each starter language.
## License
[Apache License 2.0](LICENSE). The Apache 2.0 NOTICE file is at
[`NOTICE`](NOTICE).
## Source-of-truth note
This repo is the **canonical home of the Chipzen SDK**. Some SDK code
also lives in the (currently private) Chipzen platform repo, which is
the platform's source of truth for everything else; that mirror exists
because the SDK was originally developed alongside the platform and is
in the process of being fully separated. All external development —
issues, PRs, releases — happens here.
What people ask about chipzen-sdk
What is chipzen-ai/chipzen-sdk?
+
chipzen-ai/chipzen-sdk is subagents for the Claude AI ecosystem. SDK for building and uploading bots to Chipzen — the AI poker competition platform. Wire protocol spec, client libraries, sample bots, packaging guide. It has 2 GitHub stars and its last recorded update is dated 2026-09-27.
How do I install chipzen-sdk?
+
You can install chipzen-sdk by cloning the repository (https://github.com/chipzen-ai/chipzen-sdk) or following the README instructions on GitHub. ClaudeWave also provides quick install blocks on this page.
Is chipzen-ai/chipzen-sdk safe to use?
+
Our security agent has analyzed chipzen-ai/chipzen-sdk and assigned a Trust Score of 87/100 (tier: Trusted). See the full breakdown of passed checks and flags on this page.
Who maintains chipzen-ai/chipzen-sdk?
+
chipzen-ai/chipzen-sdk is maintained by chipzen-ai. The last recorded GitHub activity is dated 2026-09-27, with 4 open issues.
Are there alternatives to chipzen-sdk?
+
Yes. On ClaudeWave you can browse similar subagents at /categories/agents, sorted by popularity or recent activity.
Deploy chipzen-sdk to your cloud
Ship this repo to production in minutes. Each platform spins up its own environment with editable env vars.
Maintain this repo? Add a badge to your README
Drop the badge into your GitHub README to show it's tracked on ClaudeWave. Each badge links back to this page and reflects the live Trust Score.
[](https://claudewave.com/repo/chipzen-ai-chipzen-sdk)<a href="https://claudewave.com/repo/chipzen-ai-chipzen-sdk"><img src="https://claudewave.com/api/badge/chipzen-ai-chipzen-sdk" alt="Featured on ClaudeWave: chipzen-ai/chipzen-sdk" width="320" height="64" /></a>More Subagents
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
The agent that grows with you
Java 面试 & 后端通用面试指南,覆盖计算机基础、数据库、分布式、高并发、系统设计与 AI 应用开发
Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without rebuilding the stack.
The agent engineering platform.
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.