{"slug": "i-measured-what-a-coding-agent-hand-off-saves-and-where-it-didn-t-help", "title": "I measured what a coding-agent hand-off saves, and where it didn't help", "summary": "A developer behind Friday, an open-source app that runs coding agents side by side, benchmarked handing a task from one agent to another against starting over, running 14 tasks including 6 in Express. Hand-offs used fewer tokens in 13 of 14 tasks (median 21%, roughly 49,000 tokens saved), cut exploration before the first edit to a median of 2 steps versus 4, and reached the first edit sooner in 12 of 14, though all 14 tasks were fixed either way and the developer makes no speed claim for Express alone. Manually pasting the first agent's full answer into the second performed about as well, so the developer frames the benefit as a one-click handover of what was asked, found, read and run rather than a token or speed advantage.", "body_md": "I build [Friday](https://github.com/splashxmoon/friday), an open-source app that runs coding agents side by side. One thing it does is hand a task from one agent to another, for example when one hits its usage limit. I wanted to know if that actually helps, or if it just sounds nice. So I measured it.\n\nClaude Code first found and explained a bug, and changed nothing. Then Codex fixed it three ways, each in its own fresh copy of the project, with one message from me:\n\nI ran 14 tasks, 6 of them in Express. Everything was measured from the session log, not from what the tools say about themselves, and checked by running the project's own tests in the folder Codex worked in.\n\n|  | Hand-off vs starting over | \n|---|---|\n| Tokens | Fewer in 13 of 14 tasks, a median of 21% (about 49,000 per hand-off). In Express alone, fewer in 6 of 6. | \n| Exploring before the first edit | A median of 2 steps against 4, fewer in 14 of 14 | \n| Speed | First edit sooner in 12 of 14, finished sooner in 11 of 14, by a median of about 7 seconds | \n| Fixed | 14 of 14 every way. In Express, all 1,288 of its tests passed after every fix. | \n\nIn Express alone the finish times were even, so I make no speed claim.\n\nPasting Claude Code's whole answer into Codex by hand did about as well: the same tokens, the same exploring, and in Express it finished faster. So I'm not claiming anything against a careful copy and paste.\n\nWhat a hand-off gives you is that handover in one click, every time: what was asked, what was found, which files were read and what was run, without you being the clipboard.\n\nI ran this on the hosted version of Friday on 13 and 14 October 2026, before the open-source app existed. The open-source app has hand-off too, but I haven't run this benchmark on it, so these numbers aren't a measurement of it.\n\nEvery run, the method and the raw data, including the runs I set aside, are on the [benchmark page](https://tryfridaywork.com/benchmarks/handoff?utm_source=devto&utm_medium=social&utm_campaign=handoff).\n\nIf you can find a flaw in how I measured, I'd like to hear it in the comments.\n\nThe open-source app is at [github.com/splashxmoon/friday](https://github.com/splashxmoon/friday): MIT, no account, on your own computer.", "url": "https://wpnews.pro/news/i-measured-what-a-coding-agent-hand-off-saves-and-where-it-didn-t-help", "canonical_source": "https://dev.to/splashxmoon/i-measured-what-a-coding-agent-hand-off-saves-and-where-it-didnt-help-1lhb", "published_at": "2026-11-02 04:38:15+00:00", "updated_at": "2026-11-02 04:52:26.563617+00:00", "lang": "en", "topics": ["ai-agents", "ai-tools", "developer-tools", "ai-research"], "entities": ["Friday", "Claude Code", "Codex", "Express", "splashxmoon"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/i-measured-what-a-coding-agent-hand-off-saves-and-where-it-didn-t-help", "markdown": "https://wpnews.pro/news/i-measured-what-a-coding-agent-hand-off-saves-and-where-it-didn-t-help.md", "text": "https://wpnews.pro/news/i-measured-what-a-coding-agent-hand-off-saves-and-where-it-didn-t-help.txt", "jsonld": "https://wpnews.pro/news/i-measured-what-a-coding-agent-hand-off-saves-and-where-it-didn-t-help.jsonld"}}