Preview deploys run PR code next to EXPO_TOKEN, a Developer robot, and that token can also publish to the production channel. A leaked token or a malicious PR could push an OTA to every production user.
Production now uses EXPO_PRODUCTION_TOKEN, a new Release Manager robot (the repo secret already exists). Previews keep EXPO_TOKEN. Once this merges, eas channel:protect production limits production publishes to Release Managers, Admins, and Owners, which leaves the preview token unable to reach production.
Order matters: merge this before protecting the channel, or production OTAs fail.
Verified: the diff only renames the secret in all 7 places plus the skip message, and vp fmt/vp lint pass. The first real check is the next production run after merge.
Model: Claude Opus 5.5 in Claude Code (T3 Code).
🤖 Generated with Claude Code
Added 1 commit
Verdict: Needs Human Approval
Macroscope's review found this PR requires human approval — Production builds and OTAs switch to a separate Expo Release Manager credential, while previews keep their existing token. Confirm the new credential’s production access and coordinate the subsequent channel protection; the first operational validation is currently deferred until after merge.
You can add or adjust custom eligibility rules. Learn more.
✅ Thread transfer remains within every enforced ceiling.
| Provider | Metric | Main baseline | This PR | Impact | PR ceiling | |
|---|---|---|---|---|---|---|
| Codex | Total thread wire | 5.0 KiB | 5.0 KiB | +29 B (+0.6%) | 6.8 KiB | ✅ |
| Codex | Thread snapshot wire | 3.8 KiB | 3.8 KiB | 0 B (0.0%) | 4.9 KiB | ✅ |
| Codex | Live turn WebSocket wire | 1.2 KiB | 1.2 KiB | +29 B (+2.3%) | 2.0 KiB | ✅ |
| Codex | Live turn WebSocket decoded | 21.5 KiB | 21.6 KiB | +41 B (+0.2%) | 24.4 KiB | ✅ |
| Codex | Live turn messages | 1 | 2 | +1 (+100.0%) | 8 | ✅ |
| Claude | Total thread wire | 5.0 KiB | 5.0 KiB | 0 B (0.0%) | 6.8 KiB | ✅ |
| Claude | Thread snapshot wire | 3.8 KiB | 3.8 KiB | 0 B (0.0%) | 4.9 KiB | ✅ |
| Claude | Live turn WebSocket wire | 1.3 KiB | 1.3 KiB | 0 B (0.0%) | 2.0 KiB | ✅ |
| Claude | Live turn WebSocket decoded | 21.9 KiB | 21.9 KiB | 0 B (0.0%) | 24.4 KiB | ✅ |
| Claude | Live turn messages | 2 | 2 | 0 (0.0%) | 8 | ✅ |
Baseline: cba3059 · PR result: 5695dfa · Source CI: success
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.
Updated in place by a trusted workflow. PR artifacts are strictly validated and never executed.
