Show HN: Hanesu – An experimental workflow layer for AI coding agents
2.25T2 sourceHacker News · workflow
Source record
Published by Hacker News · workflow (T2 source). The original is at https://github.com/jezmn/hanesu.
Pipeline notes
The summary and note below are generated by the signal pipeline — they are Beyond Desk’s reading, not quotations from the source.
SummaryShow HN introducing Hanesu, an experimental workflow layer that wraps coding agents (Codex, Claude Code, OpenCode) to structure feature, bug, and patch work with persisted task state across stages. The author notes it adds token overhead, so it targets complex tasks rather than small prompts.
Why it mattersConcrete attempt at the 'harness engineering' problem: keeping task execution state outside the chat window. Useful as one data point on multi-agent workflow structuring, though the project itself is early-stage with no traction.

Cited by
No citations on record.
