Simon Willison — Week of 2026-06-06 to 2026-06-12#

Highlight of the Week#

The standout event this week was the release of Anthropic’s massive Claude Fable 5 model, which Simon immediately leveraged as a highly capable coding partner to essentially author complex new features across his open-source ecosystem. However, the most impactful takeaway was his deep dive into the model’s terrifyingly autonomous capabilities—such as independently writing CORS servers and injecting JavaScript just to debug a CSS glitch—which served as a stark reminder of why executing AI-generated code requires strict sandboxing.

Simon Willison — 2026-06-07#

Highlight#

Simon released an early alpha of a foundational plugin that brings Claude-inspired, agentic text editing tools to the Datasette ecosystem. This creates a reliable, standardized baseline for future plugins that need to safely edit Markdown, SQL, or SVGs.

Posts#

datasette-agent-edit 0.1a0 · Source Simon released datasette-agent-edit 0.1a0 as a base plugin to simplify agentic text modifications, such as collaborative Markdown editing, updating large SQL queries, or tweaking SVG files. Noting that LLM-driven text editing is notoriously tricky to get right, he modeled the core tools—view (with line numbers), strict str_replace (which fails if the string isn’t unique), and line-based insert—directly on the published design of the Claude text editor. Rather than recreating these common patterns for every new tool, future Datasette Agent plugins can simply adapt these proven fundamentals.

Simon Willison — 2026-07-04#

Highlight#

The standout post today touches on a fascinating and slightly troubling trend in LLM tool use: state-of-the-art models like Opus 4.8 might actually be worse at interacting with custom developer tools because they are over-optimized for their proprietary, first-party environments. This highlights an emerging friction point for developers building third-party AI agents and coding harnesses.

Posts#

Better Models: Worse Tools · Source Armin Ronacher discovered that newer Anthropic models, specifically Opus 4.8 and Sonnet 5, are failing to correctly use custom tool schemas in his Pi coding harness by hallucinating extra fields, a regression not seen in older models. He theorizes this happens because these newer models are heavily trained via Reinforcement Learning to perfectly use the specific edit tools integrated directly into Claude Code. Simon points out that OpenAI models are similarly optimized on their own apply_patch mechanisms, raising the question of whether open-source and third-party harnesses will now need to maintain entirely separate edit tool implementations optimized for each specific model family.

Simon Willison — 2026-07-09#

Highlight#

The standout update today is Simon’s deep dive into the newly released GPT-5.6 family, where he unpacks OpenAI’s new API features like programmatic tool calling and analyzes their latest benchmark rivalry with Anthropic. It is a highly substantive read for developers trying to track the rapidly evolving landscape of agentic workflows and advanced API-level orchestration.

Posts#

The new GPT-5.6 family: Luna, Terra, Sol · Source OpenAI launched its GPT-5.6 flagship models in three sizes (Luna, Terra, Sol) alongside claims of superior long-running agentic performance compared to Claude Fable 5. Simon highlights the fascinating benchmark drama, noting that while Fable 5 beat GPT-5.6 Sol on SWE-Bench Pro, OpenAI recently published an article claiming that ~30% of that specific benchmark is broken. For developers, the most valuable part of the post is Simon’s exploration of new API capabilities, including a built-in multi-agent pattern, explicit prompt cache breakpoints, and “Programmatic Tool Calling” that lets models write JavaScript to orchestrate sub-tools. He also generated 18 different pelican images across the models and reasoning levels to test exact token costs.