Claim◆Article
Ruff's output provides everything a coding agent would need to fix the problems it identifies, enabling AI-assisted remediation.
The author observes that Ruff's detailed error output (location, description, suggested fix) is ideal for coding agents, and demonstrates this by having Codex and Claude Code perform the upgrades. ✦ AI generated
Brent Westbrook · Simon Willison's Weblog · 2026-07-25 · original ↗
Unsurprisingly, given Astral's new home at OpenAI, this output provides everything a coding agent would need to fix the problems. I had Codex (GPT-5.6 Sol high) upgrade LLM and sqlite-utils, and Claude Code (with Opus 5) upgrade Datasette.
Read full article ↗excerpt · fair-use quotation
- ·Ruff's output includes location, description, and suggested fix
- ·Astral's OpenAI lineage makes agent integration natural
- ·Everything a coding agent needs to fix problems is already there
- ·Codex (GPT-5.6 Sol high) upgraded LLM and sqlite-utils
- ·Claude Code (Opus 5) upgraded Datasette
- ·Both tools consumed Ruff's output to drive remediation
Around this claim
Evidence · 2
Running Ruff v0.16.0 against my three biggest Python projects found hundreds of minor issues, but comprehensive test suites made the upgrade safe.Brent Westbrook · Simon Willison's Weblog · conf 85%Ruff v0.16.0 now enables 413 rules by default, up from 59 in previous versions, dramatically increasing its default linting coverage.Brent Westbrook · Simon Willison's Weblog · conf 70%