Routi Bot’s creator claims the open-source macOS app gives each bot its own desktop and instructions, with configurable models and personal/work profile switching, potentially simplifying concurrent desktop-agent workflows.
state: expiredheat: lowuncertainty: highknownscott: lowagent-harnesses computer-use desktop-agentsnarralabswestoque
What is this?
Routi Bot is presented in a Show HN submission as an open-source macOS app whose creator says each bot gets its own desktop and instructions, with configurable AI providers or models and personal/work profile switching. The case names narralabs and westoque, but the supplied material does not establish their respective roles. None of the supplied web results directly documents Routi Bot, so its implementation, degree of desktop isolation, and ability to simplify concurrent workflows remain unverified creator claims.
Why it matters to Scott
Configurable model backends repeat Scott’s already-held Model Perishability position; dedicated bot desktops also touch Goal-World Isolation, but the creator testimony does not establish its defining per-goal lifecycle or disposable execution boundaries. The radar already tracks the adjacent separate-desktop pattern in deskwright-hidden-wayland-desktop, not Routi Bot itself; the supplied claims establish neither a consequential extension of Scott’s architecture nor a concrete reason to change his builds.
ip:concept.model-perishabilityip:concept.goal-world-isolationradar:deskwright-hidden-wayland-desktopradar:concept.computer-use
queries asked of Scott's wikis
- agent harnesses per-agent instructions execution environments
- parallel computer-use agents desktop isolation resource contention
- model-agnostic agents provider routing configuration
- personal work agent profiles context permissions separation
- macOS desktop automation agent projects
Measured heat
no measured readings yet — the hourly heat pass fills this in
How the heat travelled
no chain yet — the hourly chain pass fills this in
Evidence (2) — ⭐ canonical anchor
Interpretation history
2026-09-10T23:42:21Z
The review horizon passed without implementation evidence, independent usage reports, or a scheduled development to watch. Archive this as an unvalidated desktop-agent evaluation lead, not a disproved approach; the GitHub echo adds no independent support to the creator’s claims.
2026-09-08T22:45:32Z
No substantive delta changes the assessment: the GitHub echo repeats creator testimony rather than independently validating concurrent desktop execution or isolation. Routi Bot remains a possible evaluation target, not yet a demonstrated improvement to Scott’s workflows.
2026-09-08T22:32:29Z
grounded: known/low — Configurable model backends repeat Scott’s already-held Model Perishability position; dedicated bot desktops also touch Goal-World Isolation, but the creator te
2026-09-08T22:28:02Z
case created — The released multi-desktop harness is actionable for builders, though the announcement does not establish a security-isolation boundary.
Decision trace
- 09-11 09:42expireThe review horizon passed without implementation evidence, independent usage reports, or a scheduled development to watch. Archive this as an unvalidated desktop-agent evaluation lead, not a disproved
- 09-11 09:42alert_silentThere is no new consequential delta beyond the previously assessed announcement and no named imminent confirmation. Expiration reflects the exhausted monitoring horizon, not a reversal of the release
- 09-11 09:42alert_routeThere is no new consequential delta beyond the previously assessed announcement and no named imminent confirmation. Expiration reflects the exhausted monitoring horizon, not a reversal of the release
- 09-09 08:45repriceNo substantive delta changes the assessment: the GitHub echo repeats creator testimony rather than independently validating concurrent desktop execution or isolation. Routi Bot remains a possible eval
- 09-09 08:45alert_silentThe announcement has already been assessed, and this look adds no implementation evidence, access change, or consequential capability finding. Routine review is sufficient; there is no named imminent
- 09-09 08:45alert_routeThe announcement has already been assessed, and this look adds no implementation evidence, access change, or consequential capability finding. Routine review is sufficient; there is no named imminent
- 09-09 08:42alert_silentThe creator’s announcement establishes a new open-source macOS agent app to track, with claimed per-bot desktops, configurable models and personal/work profiles. It offers a possible future evaluation
- 09-09 08:42alert_routeThe creator’s announcement establishes a new open-source macOS agent app to track, with claimed per-bot desktops, configurable models and personal/work profiles. It offers a possible future evaluation
- 09-09 08:32groundConfigurable model backends repeat Scott’s already-held Model Perishability position; dedicated bot desktops also touch Goal-World Isolation, but the creator testimony does not establish its defining
- 09-09 08:28createThe released multi-desktop harness is actionable for builders, though the announcement does not establish a security-isolation boundary.