Willison's LLM plugin adds Claude 5 models with server-side tools
Simon Willison's llm-anthropic plugin now exposes Anthropic's latest Claude models (Fable 5, Sonnet 5, Opus 5) through the LLM command-line framework, unlocking server-side tool integration for web search, fetching, and code execution. This release bridges the gap between Anthropic's newest capabilities and the open-source LLM ecosystem, enabling developers to build agentic workflows without vendor lock-in. The timing reflects how Claude's tool-use maturity is shifting from research artifact to production infrastructure for power users and builders.
Modelwire context
Analyst takeThe real story isn't that Claude models are now available through llm-anthropic, but that Willison is deliberately routing Anthropic's latest capabilities through an open-source abstraction layer. This is a deliberate anti-lock-in move that lets developers swap Claude for GPT or other models without rewriting their agent code.
This directly operationalizes the argument Willison made on August 3rd about why devtools must be open source. He wasn't just theorizing; he's building the infrastructure that makes that freedom material. The llm-anthropic plugin is the implementation of his claim that open-source tooling becomes viable when developers can audit and modify it with AI assistance. It also connects to the datasette-apps 0.2a0 release from August 1st, which embedded agent-native debugging into data infrastructure. Both moves treat agents as first-class operators and assume developers will want to swap components without friction.
If competing model providers (OpenAI, DeepSeek, Mistral) release equivalent plugins for the llm framework within the next two months, that confirms Willison's abstraction is becoming the standard interface for agentic development. If adoption stalls because developers prefer vendor SDKs, the lock-in argument loses credibility.
Coverage we drew on
- Devtools must be open source (exe.dev) · Simon Willison
This analysis is generated by Modelwire’s editorial layer from our archive and the summary above. It is not a substitute for the original reporting. How we write it.
MentionsSimon Willison · Anthropic · Claude Fable 5 · Claude Sonnet 5 · Claude Opus 5 · LLM
Modelwire Editorial
This synthesis and analysis was prepared by the Modelwire editorial team. We use advanced language models to read, ground, and connect the day’s most significant AI developments, providing original strategic context that helps practitioners and leaders stay ahead of the frontier.
Modelwire summarizes, we don’t republish. Simon Willison originally reported this story as “llm-anthropic 0.26”. The full content lives on simonwillison.net. If you’re a publisher and want a different summarization policy for your work, see our takedown page.