Welcome to the Datasette blog

simonwillison.net·May 13, 2026

The Datasette project has launched an official blog to share upcoming announcements, created using OpenAI Codex, which includes a feature for exporting Markdown session transcripts.

The key takeaway for you is the practical application of OpenAI Codex in building a project blog, specifically leveraging its Markdown session transcript export feature. This highlights Codex's potential for enhancing workflow efficiency in AI-assisted programming, which may be directly applicable to your interests in Codex and AI productivity.

Want more content like this?

twixb tracks your favorite blogs and social media, filters by keywords, and delivers personalized key learnings — straight to your inbox.

Create Your Own →Explore Newsfeeds

More from AI Productivity

Recent stories curated alongside this one.

latent.space
[AINews] Codex Rises, Claude Meters Programmatic Usage
The launch of GPT 5.5 has sparked contrasting reactions in the AI community, with finance professionals favoring Anthropic's growth ahead of its IPO, while AI engineers show increased support for Codex due to its performance and improved limits. Meanwhile, Anthropic's recent pricing changes have faced criticism for perceived restrictions, as both companies navigate competitive dynamics in enterprise AI and developer platforms.
simonwillison.net
Quoting Boris Mann
Boris Mann argues that the phrase "11 AI agents" lacks meaning, similar to saying "11 spreadsheets" or "11 browser tabs," suggesting that the term does not convey a clear or distinct concept.
simonwillison.net
CSP Allow-list Experiment
The CSP Allow-list Experiment demonstrates how to load an application in a CSP-protected sandboxed iframe and utilize a custom `fetch()` function to intercept CSP errors, allowing users to add domains to an allow-list and refresh the page. This tool aims to enhance user experience by simplifying CSP management.
simonwillison.net
datasette 1.0a29
The release of datasette 1.0a29 includes a new utility method for creating "_r" dictionaries, improved visibility for table headers with zero rows, and fixes for bugs affecting mobile display and test crashes due to a race condition. The update enhances functionality and stability of the application.
simonwillison.net
llm 0.32a2
The release of LLM 0.32a2 introduces significant updates, notably that most reasoning-capable OpenAI models now utilize the `/v1/responses` endpoint for enhanced interleaved reasoning, allowing users to view summarized reasoning tokens in prompts. Users can opt to hide this display with specific flags if desired.

Browse all AI Productivity →

Welcome to the Datasette blog

Want more content like this?

More from AI Productivity

[AINews] Codex Rises, Claude Meters Programmatic Usage

Quoting Boris Mann

CSP Allow-list Experiment

datasette 1.0a29

llm 0.32a2