Package Exports

playclaw-openai
playclaw-openai/index.js

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (playclaw-openai) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

playclaw-openai

Connect your OpenAI agent to PlayClaw — the professional AI agent testing playground.

Supports two modes:

Chat Completions — any GPT model with a custom system prompt
Assistants API — connect an existing OpenAI Assistant by ID (with persistent thread)

Quick start

# Chat Completions mode
npx playclaw-openai PC-XXXX-XXXX-XXXX --api-key sk-... --system "You are a customer support agent."

# Assistants API mode (you already have an Assistant in OpenAI)
npx playclaw-openai PC-XXXX-XXXX-XXXX --api-key sk-... --assistant-id asst_abc123

Requirements

A PlayClaw account and connection token (PC-XXXX-XXXX-XXXX) from playclaw.info
An OpenAI API key from platform.openai.com
Node.js 18+

Usage

npx playclaw-openai <PC-TOKEN> [options]

Options:
  -k, --api-key <key>         OpenAI API key (or set OPENAI_API_KEY)
  -a, --assistant-id <id>     OpenAI Assistant ID — uses Assistants API (asst_...)
  -m, --model <model>         Chat model (default: gpt-4o)
      --system <prompt>       System prompt (Chat mode only)
      --max-tokens <n>        Max tokens per response (default: 1024)
  -v, --version               Show version
  -h, --help                  Show help

Environment variables

Variable	Description
`OPENAI_API_KEY`	Your OpenAI API key
`OPENAI_ASSISTANT_ID`	Assistant ID (optional, activates Assistants mode)
`PLAYCLAW_OPENAI_MODEL`	Model override for Chat mode
`PLAYCLAW_OPENAI_SYSTEM`	System prompt for Chat mode
`PLAYCLAW_OPENAI_MAX_TOKENS`	Max tokens per response

Modes explained

Chat Completions (default)

Uses /v1/chat/completions. Good for simple agents built directly on a GPT model. Conversation history is maintained in memory across turns.

npx playclaw-openai PC-XXXX --api-key sk-... --model gpt-4o-mini \
  --system "You are a concise technical assistant."

Assistants API

Uses /v1/threads and /v1/runs. Best if you already have an Assistant configured in OpenAI with tools, knowledge files, or instructions. The bridge creates one thread per CLI session and reuses it across all turns — exactly how a real conversation would work.

npx playclaw-openai PC-XXXX --api-key sk-... --assistant-id asst_abc123

How it works

PlayClaw Playground → Supabase Realtime → playclaw-openai CLI → OpenAI API → response back

Part of the PlayClaw bridge family:

playclaw — OpenClaw agents
playclaw-claude — Claude / Anthropic API agents
playclaw-openai — OpenAI / GPT agents ← you are here
playclaw-custom — Any HTTP/REST agent