fix(devices): apply settings to selected environments

[rynfar]

Device host edits previously changed only the representative environment even when several environments were selected in Settings. Adding, editing, and removing a host now updates each selected environment's own list, retains each environment's identity-file path unless the user edits that field, and reports partial or offline failures by environment. Editing onto another host's destination is rejected instead of creating duplicate targets. The device hub toggle applies to selected environments, while project and checkout scopes keep the existing project-level agent-device permission switch. Status and tool inspection remain tied to the representative environment.

This adapts upstream 66e39ca2aabde054bc50312a9c34f05dbd1f6f9e while preserving Pylon's host retry, local tool status, and selected-environment connection checks. #731 and #766 have landed; the current PR diff is limited to the nine settings/docs files for this behavior.

Validation on the reconciled tree: 53 focused settings and host-check tests, web typecheck, scoped lint and format, and a frozen offline install passed. The later base merge changed no file content. Browser/device interaction remains unverified. All reported final-head CI checks pass at bf23678a777a17c8d3e4c8d8588fddcf7cf19817, including web checks, server tests, and Linux/Windows packaged smoke. Browser/device UI evidence is still pending before merge.

Implemented and reviewed by GPT-6 Sol in the Codex/Pylon harness. Refs #736.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

1 Skipped Deployment

Project	Deployment	Actions	Updated
pylon-marketing	Ignored	Preview	Sep 24, 2026 11:10am UTC

[github-actions]

Thread transfer impact

✅ Thread transfer remains within every enforced ceiling.

Provider	Metric	Main baseline	This PR	Impact	PR ceiling
Codex	Total thread wire	14.0 KiB	13.9 KiB	−7 B (−0.0%)	15.1 KiB	✅
Codex	Thread snapshot wire	7.2 KiB	7.2 KiB	−5 B (−0.1%)	7.3 KiB	✅
Codex	Live turn WebSocket wire	6.7 KiB	6.7 KiB	−2 B (−0.0%)	7.8 KiB	✅
Codex	Live turn WebSocket decoded	58.0 KiB	58.0 KiB	0 B (0.0%)	66.4 KiB	✅
Codex	Live turn messages	9	9	0 (0.0%)	21	✅
Claude	Total thread wire	14.0 KiB	14.0 KiB	−8 B (−0.1%)	15.1 KiB	✅
Claude	Thread snapshot wire	7.2 KiB	7.2 KiB	−6 B (−0.1%)	7.3 KiB	✅
Claude	Live turn WebSocket wire	6.7 KiB	6.7 KiB	−2 B (−0.0%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	58.9 KiB	58.9 KiB	0 B (0.0%)	66.4 KiB	✅
Claude	Live turn messages	9	9	0 (0.0%)	21	✅

Baseline: 89e17c6 · PR result: bf23678 · Source CI: success

Scenario and decoded snapshot size

10 historical turns, 5 command tools per turn, 878.9 KiB retained MCP result per historical turn, and a 1.05 MiB retained result in the measured turn.

• Codex decoded thread snapshot: 115.6 KiB
• Claude decoded thread snapshot: 116.3 KiB

Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.

[vercel]

Deployment failed for project pylon-marketing with the following error:

Resource is limited - try again in 24 hours (more than 100, code: "api-deployments-free-per-day").

Learn More: https://vercel.com/pylon-code?upgradeToPro=build-rate-limit

[rynfar]

Reconciled this stacked PR onto current pylon and the refreshed #731 prerequisite at 4ad09b330064b97bf4cb46debcefd2ca52a5e141. The combined settings tree passes 53 focused settings/host-check tests, web typecheck, scoped lint/format, and a frozen offline install. Review found one real collision path: editing a host onto another host's destination could create duplicate targets in a selected environment. The planner now rejects that destination and a regression covers it.

Final-head CI is running. Browser/device interaction is still unverified. Merge remains subject to #731 landing and all checks green; Vercel currently reports the account's daily deployment quota on this head.

[rynfar]

Isolated browser QA on final head bf23678a777a17c8d3e4c8d8588fddcf7cf19817: paired a second disposable backend, selected All environments, added qa@host.invalid, and confirmed the host appeared in both environments. Selected the second environment separately and confirmed its host entry remained visible. The browser console showed no errors. The first image shows the multi-environment add form; the second shows both saved entries and the environment selector.

These backends used disposable state only. The placeholder SSH target was saved but never connected to.