Web Search and Fetch for OpenClaw

Ollama now ships with web search and web fetch plugins for OpenClaw, allowing local and cloud models to search the web for the latest content and news. The web fetch capability extracts readable content for processing, though JavaScript execution is not supported in this implementation.

To use web search with local models in OpenClaw, sign in with ollama signin, then launch with ollama launch openclaw. If you already have OpenClaw configured, you can install the web search plugin directly:

openclaw plugins install @ollama/openclaw-web-search

Headless Mode for Automation

The ollama launch command now supports non-interactive (headless) mode, making it ideal for containerized deployments and automation:

• Docker/containers: Spin up integrations as pipeline steps to run evaluations, test prompts, or validate model behavior
• CI/CD: Generate code reviews, security checks, and other automated tasks
• Scripts and automation: Kick off automated workflows with Ollama and Claude

To run in headless mode:

• --model parameter is required
• Use --yes flag to auto-pull models and skip interactive selectors

Example: ollama launch claude --model kimi-k2.5:cloud --yes -- -p "how does this repository work?"

Additional Improvements

• ollama launch openclaw now uses official Ollama authentication and model provider integration
• Improvements to Ollama's benchmarking tool in ./cmd/bench
• ollama launch openclaw now skips --install-daemon when systemd is unavailable