pluginintermediateOpenCLAW

OpenCLAW v2026.3 Update: New Providers, Plugin Enhancements, and ClawHub

The OpenCLAW framework update improves its flexibility by adding support for more AI model providers like Ollama and expanding its plugin system. This allows businesses to build more customized and powerful personal AI assistants using a wider range of tools and models.

AI SETUP PROMPT

Paste into Claude Code or Codex CLI — it will scan your project and set everything up

# Install & Configure: OpenCLAW v2026.3 Update: New Providers, Plugin Enhancements, and ClawHub

## What This Is
The OpenCLAW framework update improves its flexibility by adding support for more AI model providers like Ollama and expanding its plugin system. This allows businesses to build more customized and powerful personal AI assistants using a wider range of tools and models.

Source: https://github.com/orgs/openclaw/repositories

## Before You Start

Scan my workspace and analyze:
- The project language, framework, and directory structure
- Existing agent configuration (check for .claude/, .codex/, CLAUDE.md, settings.json, commands/, skills/ directories)
- Whether this repository or a similar tool is already cloned or installed

Then ask me before proceeding:
1. Do I need to configure any service credentials for this plugin (database, API keys, etc.)?
2. Should this be project-scoped or global?

## What to Implement

This is an **Agent Plugin** — typically an MCP server or capability bundle.

- Install any required npm/pip packages from the source repository
- Register the MCP server in `.claude/settings.json` under the `mcpServers` key with the correct command and args
- Configure any required environment variables or credentials
- Verify the server starts correctly by checking its expected output

## Additional Context

- Scan the local project for an existing OpenCLAW installation. If found, execute `claw update --to v2026.3.13` to upgrade. If not found, clone the core repository with `git clone https://github.com/openclaw/openclaw.git`.
- Access the official skill directory repository at `https://github.com/openclaw/clawhub` to identify new or updated provider plugins (e.g., Ollama, vLLM, OpenRouter) and skill bundles relevant to the user's project goals.
- Based on the identified plugins from ClawHub, execute the appropriate `claw install <plugin_name>` command. If a new provider plugin requires an API key, check the user's environment variables for it and prompt them to add one if missing.

## Guidelines

- Adapt everything to my existing project — do not assume a specific stack or directory layout
- Review any fetched code for safety before installing or executing it
- After setup, run a quick verification and show me a summary of exactly what was installed, where, and how to use it

2,337 charactersCompatible with Claude Code & Codex CLI

MANUAL SETUP STEPS

01Scan the local project for an existing OpenCLAW installation. If found, execute `claw update --to v2026.3.13` to upgrade. If not found, clone the core repository with `git clone https://github.com/openclaw/openclaw.git`.
02Access the official skill directory repository at `https://github.com/openclaw/clawhub` to identify new or updated provider plugins (e.g., Ollama, vLLM, OpenRouter) and skill bundles relevant to the user's project goals.
03Based on the identified plugins from ClawHub, execute the appropriate `claw install <plugin_name>` command. If a new provider plugin requires an API key, check the user's environment variables for it and prompt them to add one if missing.

FIELD OPERATIONS

Local-First RAG Agent

Create a research assistant that uses the new Ollama provider plugin to run a local language model (like Llama 3). The agent will use OpenCLAW's file system skills to read local documents, generate embeddings, and answer questions without sending any data to the cloud.

Cross-Provider Code Generation Bot

Build a bot that leverages multiple AI provider plugins via bundle support (e.g., OpenRouter, Claude, Codex). The bot will take a complex coding task, send it to multiple models simultaneously, and use a custom 'judge' skill to compare the outputs for correctness and style, presenting the best solution.

STRATEGIC APPLICATIONS