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 — Week of 2026-06-12 to 2026-06-18#

Highlight of the Week#

The most impactful release this week is the launch of datasette-apps, a major new plugin that allows developers to run self-contained, sandboxed HTML and JavaScript applications directly against a persistent Datasette backend. It brilliantly merges Simon’s ongoing experiments with AI-generated “vibe-coded” single-file tools and robust security architectures, pushing Datasette from a read-only publishing platform into a comprehensive ecosystem for building interfaces over data.

Simon Willison — Week of 2026-06-18 to 2026-06-25#

Highlight of the Week#

This week’s absolute standout is the launch of the datasette-apps plugin, which fundamentally transforms how we build micro-applications over local databases. By utilizing tightly constrained iframe sandboxes and Content-Security-Policy headers, developers and LLMs alike can safely run custom HTML/JS interfaces against a persistent Datasette backend. It brilliantly merges Simon’s ongoing experiments with AI-assisted “vibe coding” and robust security architectures into a core ecosystem feature, effectively bridging the gap between Claude Artifacts and secure data environments.

Simon Willison — 2026-07-13#

Highlight#

DOOMQL stands out as a wonderfully unreasonable experiment—running a Doom engine entirely in SQLite. It perfectly highlights the creative potential of AI-assisted programming when combined with Simon’s ecosystem, as he used Claude to instantly build a live-updating companion minimap using his new Datasette Apps plugin.

Posts#

DOOMQL · Source Peter Gostev used GPT-5.6 Sol to build a functional Doom-like game where SQLite acts as the game engine, handling everything from collision to a recursive CTE ray tracer for rendering. Simon took this a step further by using Claude Fable 5 and his Datasette Apps plugin to quickly generate a live-updating HTML and JavaScript minimap that reflects the game state in the browser while playing in the terminal. It is a brilliant showcase of using LLMs to push small sharp tools to their absolute limits.

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.

Simon Willison — 2026-07-08#

Highlight#

Jarred Sumner’s post on rewriting Bun from Zig to Rust using AI agents is an incredible showcase of how frontier LLMs are upending the old Spolsky rule of “never rewrite from scratch”. It is a masterclass in agentic engineering, utilizing dynamic workflows and a TypeScript conformance suite to successfully port millions of lines of code.

Posts#

Rewriting Bun in Rust · Source Jarred Sumner details the agentic engineering process of rewriting Bun from Zig to Rust to solve complex memory management issues. Using a language-independent TypeScript test suite as a conformance suite, an agent harness powered by Claude Mythos/Fable automated the massive code translation. The sheer scale of the project required 5.9 billion input tokens—around $165,000 at API pricing—proving that coordinated parallel agents fundamentally change the calculus of ground-up software rewrites.

Engineering Reads — 2026-04-04#

The Big Idea#

Raw LLM intelligence is no longer the primary bottleneck for AI-assisted development; the real engineering challenge is building the system scaffolding—memory, tool execution, and repository context—that turns a stateless model into an effective, autonomous coding agent.

Deep Reads#

[Components of A Coding Agent] · Sebastian Raschka · Sebastian Raschka Magazine The core insight of this piece is that an LLM alone is just a stateless text generator; to do useful software engineering, it needs a surrounding agentic architecture. Raschka details the necessary scaffolding: equipping the model with tool use, stateful memory, and deep repository context. The technical mechanism relies on building an environment where the model can fetch file structures, execute commands, and persist state across conversational turns rather than just blindly emitting isolated code snippets. The tradeoff here is a steep increase in system complexity—managing context windows, handling tool execution failures, and maintaining state transitions is often much harder than prompting the model itself. Systems engineers and developers building AI integrations should read this to understand the practical anatomy of modern autonomous developer tools.

Simon Willison — 2026-04-05#

Highlight#

Simon highlights a deep-dive post by Lalit Maganti on the realities of “agentic engineering” when building a robust SQLite parser. The piece beautifully articulates a crucial lesson for our space: while AI is incredible at plowing through tedious low-level implementation details, it struggles significantly with high-level design and architectural decisions where there isn’t an objectively right answer.

Posts#

Eight years of wanting, three months of building with AI Simon shares a standout piece of long-form writing by Lalit Maganti on the process of building syntaqlite, a parser and formatter for SQLite. Claude Code was instrumental in overcoming the initial hurdle of implementing 400+ tedious grammar rules, allowing Lalit to rapidly vibe-code a working prototype. However, the post cautions that relying on AI for architectural design led to deferred decisions and a confusing codebase, ultimately requiring a complete rewrite with more human-in-the-loop decision making. The core takeaway is that while AI excels at tasks with objectively checkable answers, it remains weak at subjective design and system architecture.

Simon Willison — 2026-04-07#

Highlight#

Anthropic’s decision to restrict access to their new Claude Mythos model underscores a massive, sudden shift in AI capabilities. It is a fascinating look at an industry-wide reckoning as open-source maintainers transition from dealing with “AI slop” to facing a tsunami of highly accurate, sophisticated vulnerability reports.

Posts#

[Anthropic’s Project Glasswing - restricting Claude Mythos to security researchers - sounds necessary to me] · Source Anthropic has delayed the general release of Claude Mythos, a general-purpose model similar to Claude Opus 4.6, opting instead to limit access to trusted partners under “Project Glasswing” so they can patch foundational internet systems. Simon digs into the context, tracking how credible security professionals are warning about the ability of frontier LLMs to chain multiple minor vulnerabilities into sophisticated exploits. He even uses git blame to independently verify a 27-year-old OpenBSD kernel bug discovered by the model. He concludes that delaying the release until new safeguards are built, while providing $100M in credits to defenders, is a highly reasonable trade-off.

Simon Willison — 2026-04-08#

Highlight#

The most substantial piece today is a deep-dive into Meta’s new Muse Spark model and its chat harness, where Simon successfully extracts the platform’s system tool definitions via direct prompting. His exploration of Meta’s built-in Python Code Interpreter and visual_grounding capabilities highlights a powerful, sandbox-driven approach to combining generative AI with programmatic image analysis and exact object localization.

Posts#

Meta’s new model is Muse Spark, and meta.ai chat has some interesting tools Meta has launched Muse Spark, a new hosted model currently accessible as a private API preview and directly via the meta.ai chat interface. By simply asking the chat harness to list its internal tools and their exact parameters, Simon documented 16 different built-in tools. Standouts include a Python Code Interpreter (container.python_execution) running Python 3.9 and SQLite 3.34.1, mechanisms for creating web artifacts, and a highly capable container.visual_grounding tool. He ran hands-on experiments generating images of a raccoon wearing trash, then used the platform’s Python sandbox and grounding tools to extract precise, nested bounding boxes and perform object counts (like counting whiskers or his classic pelicans). Although the model is closed for now, infrastructure scaling and comments from Alexandr Wang suggest future versions could be open-sourced.