Datasette

By running Python ASGI web applications entirely in the browser using Pyodide and a dedicated service worker, this project intercepts all same-origin requests under `/app/` and executes them against the Python app via the ASGI protocol—removing the need for a backend server except for static files. The mechanism is demonstrated with both a FastAPI demo and the full Datasette app, confirming its generality across ASGI apps.

Datasette 1.0a31 新增支持执行写入 SQL 和保存存储查询功能，提升数据库交互能力。

Datasette Agent is the first LLM-powered AI assistant for Datasette, enabling conversational data querying and chart generation via Gemini 3.1 Flash-Lite model with plugin extensibility.

Datasette Agent 0.1a4 integrates with a new JS plugin hook to enable launching an AI chat interface via shortcut key for enhanced data analysis interaction.

Datasette 1.0a30 introduces a customizable "Jump to" menu for improved navigation and plugin extensibility.

Datasette 1.0a30 introduces a "jump to" menu accessible via the "/" shortcut, enabling quick navigation to databases, tables, or queries, and supports plugin extensions.

The first alpha version of Datasette Agent has been released, which is a conversational AI assistant for Datasette that can answer questions about data in SQLite databases and supports plugin extensions. This tool introduces natural language query capabilities into SQLite data analysis scenarios, providing new interaction methods for data exploration.

Datasette Agent is officially released as the first extensible AI assistant for Datasette, built on over three years of LLM Python library development experience, providing intelligent support for data querying and analysis.

Simon Willison released datasette-ip-rate-limit 0.1a0 to provide IP-based rate limiting for Datasette requests.

datasette 1.0a29 released with new utility methods and bug fixes.

Using LLM in the shebang line of a script is an innovative practice that enables more flexible command execution and tool integration.

LLM can be used directly in the shebang line of a script, supporting tool calls and complex template processing, but use with caution.

datasette-fixtures 0.1a0 provides a way to populate test databases, streamlining plugin development workflows.

The latest datasette-agent plugin update introduces a form in the menu's empty state to initiate agent conversations, enhancing user interaction.

datasette-llm 0.1a7 releases a standardized LLM configuration layer for the Datasette ecosystem, enabling plugins to share default model parameters (e.g., model name, temperature) to improve reusability and consistency of AI features.

Simon Willison released datasette-referrer-policy 0.1 to solve OpenStreetMap tile loading failures caused by Datasette's default 'no-referrer' policy, enabling configurable HTTP header control without breaking defaults.

datasette-agent 0.1a3 is released, but the article focuses heavily on ads and sponsorship with minimal technical details.

Simon Willison announced the alpha release of datasette-agent-charts 0.1a2, but the article contains minimal technical details and is dominated by advertising and sponsorship content.

Simon Willison's May newsletter summary covers AI cost increases and Anthropic's performance, but lacks technical depth, unsuitable for deep engineer reading.

datasette 1.0a32 is a minor bugfix release of an open-source multi-tool for exploring and publishing data, fixing bugs in `INSERT ... RETURNING` queries via the new `/db/-/execute-write` endpoint and several `base_url` issues discovered during Service Workers experimentation.

datasette-llm-accountant 0.1a4 is a minor tool for tracking LLM usage costs in Datasette, with only a bug fix for response chain tracking and no new features or architectural innovation, making it a low-information release note.

Simon Willison released the datasette-agent-sprites 0.1a0 plugin, enabling command execution in Fly Sprites sandboxes.

This tweet is merely a repost of a blog update with no substantive technical content or depth, offering limited informational value.