The Archive

Simon Willison argues LLMs make open-source dev tools more practical by lowering code-reading barriers through AI-assisted comprehension.

Simon Willison releases condense-json 1.0, a utility library for compressing JSON by replacing repeated strings with references.

Simon Willison's June 2026 newsletter roundup covering model releases (GPT-5.6, Claude Opus 5, DeepSeek-V4), open letters, and accidental cyberattacks by OpenAI and Anthropic test models.

MCP 2.0 (stateless Model Context Protocol) launches with significant spec changes; Simon Willison revives mcp-explorer and datasette-mcp tooling.

Simon Willison releases llm-mcp-client 0.1a0, enabling Model Context Protocol integration for LLM tools.

Podcast discussion on open-weight model competitiveness, cybersecurity risks, and AI leadership policy letters signed by industry leaders.

Simon Willison releases smevals, an open eval framework for benchmarking models, prompts, and inference harnesses across configurations.

Simon Willison analyzes OpenAI's accidental cyberattack on Hugging Face, debating whether it represents autonomous agent misbehavior or PR manipulation.

Simon Willison summarizes Nik Suresh's critique of AI-driven decision-making at large enterprises, highlighting executives making AI strategy without hands-on tool experience.

Simon Willison built an interactive SQLite query plan explainer using Claude to generate explanations of EXPLAIN output in the browser via WebAssembly.

Simon Willison notes recent activity on Quixote, a 21-year-old Python web framework.

Simon Willison comments on Kimi K3's refusal to leak its system prompt, noting the model's polite deflection.

Simon Willison releases tool to detect common linguistic patterns in LLM-generated text.

Simon Willison compiled Go's mermaid-ascii library to WebAssembly for ASCII diagram rendering with color support.

Simon Willison ports Grok's Rust Mermaid-to-Unicode renderer to WebAssembly for browser use via Claude Code.

Simon Willison documents a data exfiltration vulnerability in Claude's web_fetch tool that exploits interaction between private memories and URL-based attacks.

Simon Willison shares a caching technique for uvx Python tools in GitHub Actions using UV_EXCLUDE_NEWER environment variable.

Simon Willison documents productivity spike in Datasette project correlating with Opus 4.8 and GPT-5.5 releases via GitHub commit frequency analysis.

Simon Willison examines DRI (Directly Responsible Individual) concept from Apple/GitLab in context of LLM agents, arguing humans must retain accountability.

llm-meta-ai 0.1 released, enabling CLI access to Meta's muse-spark-1.1 model via Simon Willison's llm tool.

Simon Willison reports Anthropic's Claude Code team recommends letting Fable/Opus apply autonomous judgment on tasks like testing rather than prescribing workflows.

Simon Willison's June 2026 newsletter covers Claude Fable 5, GPT-5.6, GLM-5.2 open weights, and US export restrictions.

Simon Willison releases llm-coding-agent 0.1a0, a Python agent framework for LLM-based code generation with file and command execution tools.

Simon Willison uses DSPy to evaluate and optimize SQL system prompts in Datasette Agent, a tool enabling agents to execute read-only database queries.

Simon Willison reviews an interactive quiz that classifies users into 30 AI ethics archetypes based on 29 questions.

Simon Willison releases HTML table extractor tool supporting conversion to Markdown, CSV, JSON, and other formats.

Simon Willison uses Claude to reconstruct a 1983 Usborne computer game (Mad House) as interactive JavaScript from a PDF.

Simon Willison releases Datasette Agent, a conversational AI assistant for querying structured data with chart generation capabilities.

Simon Willison reviews Google I/O announcements: Gemini 3.5 Flash GA and Gemini Spark (OpenClaw competitor) in preview, emphasizing lack of hands-on availability for most launches.

Simon Willison summarizes six months of LLM developments (Nov 2025 onward) including a November inflection point and rapid shifts in top model performance, especially for coding.