Skip to content

Instantly share code, notes, and snippets.

@simonw
Created June 9, 2026 23:29
Show Gist options
  • Select an option

  • Save simonw/8d05ce836d44bd8543055c9614f7c478 to your computer and use it in GitHub Desktop.

Select an option

Save simonw/8d05ce836d44bd8543055c9614f7c478 to your computer and use it in GitHub Desktop.

2026-06-09T23:23:00 conversation: 01ktqb6zs46fs52eg6sytd3y0m id: 01ktqb4jjss14ytd9zyymw605c

Model: gpt-5.5

Prompt

List all of Simon Willion's open source projects, most recent first, each with a rough date of when they were first released

Reasoning

Clarifying project details

I'm trying to decide which projects to include, and it seems I should focus only on prominent ones rather than plugins since the user didn't specify repositories. I don't want to overpromise, so I'll clarify that I can't list every repo without a live GitHub query.

For the projects, I need to gather accurate release dates: Django was first released in July 2005, while Datasette launched in November 2017. I’ll gather more release date details!Reviewing Datasette plugins

I’m looking into a list of various Datasette plugins and their development timelines. There's a range of plugins from as early as 2018, like datasette-auth-github and others that were introduced in 2020. It seems the ecosystem has grown over time. I’m thinking about how to summarize this without listing every individual plugin, which might be overwhelming. Maybe it’s best to acknowledge that there isn't a complete catalog available while exploring Simon Willison's projects page for dates and entries.Reviewing memory of tools

I'm reflecting on various tools and projects, noting their descriptions and development dates. For instance, "LLM" is a CLI utility and Python library for working with Large Language Models since 2023, while "Datasette" has been around since 2017 for exploring and publishing data. Then there's "sqlite-utils," launched in 2018, and the Django web framework dating back to 2005. I’m also pondering some older projects like "django-people" and others that might not be open source. This list helps me track these tools efficiently.Organizing OSS Projects Timeline

I’m working on compiling a timeline of major standalone open-source software (OSS) projects. I want to sort everything by date, aiming for 25 entries. My proposed dates are from 2005 to 2024. I need to clarify certain projects, like "yaml-to-sqlite" in late 2019 and the potential inclusion of "airtable-export" from 2020. I’m also not sure about some details, like whether "djangopeople.net" qualifies. Let's get this timeline sorted!Refining OSS Projects List

I need to confirm details about certain projects, like "airtable-export" from 2020, which was done by Simon Willison, and "swarm-to-sqlite" for Foursquare Swarm, released also in 2020. "twitter-to-sqlite" was first released in 2019. I should correct a name to "Willison" and include notes about "dogsheep-beta" and "dogsheep-photos," both possibly from 2020. In the final list, I want to clarify that not every small repo needs to be included. Also, offering to list repos via the GitHub API is an option, though the user didn't request code.

Response

Assuming you mean Simon Willison: below is a most-recent-first list of his major/recognizable open-source projects and project families. “All” is hard to make exact because he has many small repos, Datasette plugins, LLM plugins, demos, and experiments; dates are rough first-public-release/announcement dates.

First released Project Notes
2024 files-to-prompt CLI tool for gathering files into prompts for LLMs.
2023 strip-tags CLI/Python tool for stripping HTML tags, useful for preparing text for LLMs.
2023 symbex CLI tool for extracting/searching symbols from Python code.
2023 LLM CLI and Python library for interacting with large language models, with a plugin ecosystem.
2023 ttok CLI tool for counting/truncating text by LLM token counts.
2023 openai-to-sqlite Tooling for exporting/logging OpenAI-related data to SQLite.
2022 shot-scraper CLI tool for taking automated screenshots of web pages using Playwright.
2022 Datasette Lite Runs Datasette entirely in the browser using WebAssembly/Pyodide.
2022 django-http-debug Django app for logging/debugging HTTP requests and responses.
2021 django-sql-dashboard Django app for building dashboards from SQL queries.
2021 s3-credentials CLI tool for creating tightly scoped AWS S3 credentials.
2021 Datasette Desktop / Datasette App Desktop packaging/experiments for running Datasette locally.
2020 Dogsheep Personal analytics ecosystem: import your own data into SQLite/Datasette.
2020 dogsheep-beta Search/indexing interface for Dogsheep data.
2020 git-scraper Pattern/tooling for scraping data into Git history over time.
2020 healthkit-to-sqlite Imports Apple HealthKit exports into SQLite.
2020 google-takeout-to-sqlite Imports selected Google Takeout data into SQLite.
2020 pocket-to-sqlite Imports Pocket saved-items data into SQLite.
2020 hacker-news-to-sqlite Imports Hacker News data into SQLite.
2020 swarm-to-sqlite Imports Foursquare/Swarm check-ins into SQLite.
2020 inaturalist-to-sqlite Imports iNaturalist observations into SQLite.
2020 airtable-export Exports Airtable data, often for use with Datasette/SQLite workflows.
2019 github-to-sqlite Imports GitHub API data into SQLite.
2019 twitter-to-sqlite Imports Twitter/X data into SQLite.
2019 yaml-to-sqlite Converts YAML files into SQLite databases.
2018–2019 sqlite-utils Python library and CLI for creating, querying, transforming, and manipulating SQLite databases.
2018 db-to-sqlite Imports data from other database engines into SQLite.
2018 onward Datasette plugins Large plugin ecosystem, including plugins for auth, maps, charts, publishing, rendering, search, and more.
2017 Datasette Tool for exploring and publishing data as interactive SQLite-backed websites/APIs.
2017 csvs-to-sqlite Converts CSV files into SQLite databases; closely tied to early Datasette workflows.
2005 Django Major Python web framework; Simon Willison was one of its original creators.

So, in broad terms, his open-source work clusters around Django, Datasette/SQLite/data tooling, Dogsheep personal analytics, and more recently LLM command-line tooling.

Token usage

32 input, 4,026 output, {"input_tokens_details": {"cached_tokens": 0}, "output_tokens_details": {"reasoning_tokens": 3105}}

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment