Codex ports a Python HTML parser to JavaScript
Willison supplied an existing parser and an independent compatibility suite, reviewed an API plan, then let Codex implement a JavaScript port through repeated tests and commits.
simonwillison.net
Suggested prompt
Suggested prompt written by Agents In Use, not the source author. Replace bracketed inputs and review tool permissions first.
Port [library] to [language] in a separate directory. Inspect its license, API and independent compatibility suite. Write a plan and implement one smoke case first. Continue through the suite, preserving skipped and failing cases in the report. Stop before release.
Reach people exploring AI agents
See placements →Evidence at a glance
Sourced account- Source author
- Simon Willison
- Original account
- Read the source ↗
- Published
- Source link checked
- Reported result
- Willison reports a dependency-free JavaScript library passing thousands of compatibility tests; the session summary also lists skipped cases.
- Human involvement
- Willison reviewed the specification and smoke milestone, monitored commits and resumed the run after a usage limit.
A linked account supports attribution. It does not mean we independently reproduced its result.
Report an issue with this case
Open correction emailOpens your email app with the case link included. Nothing is sent automatically.
How the agent did it
- Read the Python parser and compatibility tests and prepare an API specification.
- Implement a small working parser before expanding through the planned milestones.
- Run the compatibility suite repeatedly and commit progress, then add a browser playground.
More details
About this account
Codex worked through implementation milestones and test failures between a small number of follow-up instructions.
Firsthand technical write-up links code and tests. Compatibility is a reported outcome, not a guarantee of production suitability.
Credit: Simon Willison
Source published
Added to Agents In Use
Source checked
Tools and topics
Codex CLI · GPT-5.2 · html5lib-tests · Git · GitHub Actions
Code · Testing
Editorial completeness: 100/100. Measures documented fields, not agent performance or verified outcomes.