feat(ci): restore anonymous macOS PR previews

[rynfar]

Opted-in macOS PR builds currently require GitHub sign-in to download. Restore anonymous Apple Silicon DMGs on Pylon's rolling desktop-preview prerelease, with direct links and SHA-256 checksums in the PR's bot comment.

The PR build has read-only permissions and keeps Pylon's unsigned -pr. packaging and Alpha identity. A separate publisher executes trusted pylon scripts, checks the current same-repository PR/head/build attempt/label, and treats the downloaded artifact as opaque bytes. Publication and cleanup share a per-PR lock; new assets upload before older ones are pruned. API outages fail visibly, and close/opt-out cleanup affects only that PR's validated Pylon filenames. Production releases and update feeds remain in pylon-code/pylon-releases; preview packaging already disables updates.

This reopens DEF-7 in #526. It restores the behavior adopted in #111 and repaired by 9d112329eb199af97b22025b14b0c0ea1ccea482, then reverted in #113. Original upstream source: 33b650a5b3b27382b35d2182dec6b22438c3da56 (#8243). The trusted publisher split is a Pylon adaptation.

Validation:

• 62 passing focused Node tests across preview publication and existing nightly source selection, including stale runs/attempts, filename attacks, same-PR cleanup, API errors, upload interruption, initial-release races, and retry behavior.
• Actionlint 1.7.12 passes both preview workflows and the affected CI workflow.
• Vite+ 0.3.0 focused formatting/lint and the normal pre-commit hook pass.

Operations and recovery are documented in docs/operations/desktop-previews.md. No label was created/applied, preview build dispatched, release published, or app installed during this change. Merging alone does not publish a preview; a maintainer must opt a trusted PR in with preview:mac. The GitHub-hosted publication and anonymous download path remain to be exercised on the first authorized preview.

Model: GPT-6. Harness: Codex.

---

^{Need help on this PR? Tag @codesmith-bot with what you need. Autofix is disabled.}

[vercel]

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

Project	Deployment	Actions	Updated
pylon-marketing	Ready	Preview	Sep 12, 2026 4:27pm 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	13.9 KiB	13.9 KiB	+14 B (+0.1%)	15.1 KiB	✅
Codex	Thread snapshot wire	7.2 KiB	7.2 KiB	−2 B (−0.0%)	7.3 KiB	✅
Codex	Live turn WebSocket wire	6.7 KiB	6.7 KiB	+16 B (+0.2%)	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	8	8	0 (0.0%)	21	✅
Claude	Total thread wire	13.9 KiB	13.9 KiB	+7 B (+0.0%)	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	+1 B (+0.0%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	58.8 KiB	58.8 KiB	0 B (0.0%)	66.4 KiB	✅
Claude	Live turn messages	8	8	0 (0.0%)	21	✅

Baseline: d6d2f8e · PR result: f05bf60 · 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: 115.7 KiB
• Claude decoded thread snapshot: 116.4 KiB

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