Web Search and Fetch for OpenClaw

Ollama now ships with web search and web fetch plugins for OpenClaw, allowing both local and cloud-based models to search the web for current content and news. The fetch capability extracts readable content without executing JavaScript, providing safe content retrieval for model processing.

To enable this feature with local models, authenticate with ollama signin, then launch OpenClaw with ollama launch openclaw. If you already have OpenClaw configured, install the web search plugin directly:

openclaw plugins install @ollama/openclaw-web-search

Non-Interactive Headless Mode

The ollama launch command now supports non-interactive mode for automation scenarios. Key features include:

• Docker/Containers: Spin up integrations as pipeline steps to run evaluations, test prompts, or validate model behavior during builds
• CI/CD Integration: Generate code reviews, security checks, and other automated tasks within your CI pipeline
• Scripts & Automation: Trigger automated workflows with Ollama and Claude code

Required flags for headless operation:

• --model must be specified
• --yes flag auto-pulls models and skips interactive selectors

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

Additional Improvements

• Improved Ollama authentication and model provider integration for OpenClaw
• Enhanced benchmarking tools in ./cmd/bench
• Fixed systemd detection to skip --install-daemon when unavailable