docs(upstream): close the frozen integration cycle and advance the cursor

[rynfar]

The frozen T3 integration cycle is fully classified and its last ports are merged. Advance reviewed-through from beae2147a9487ec47ac992319f2216914b4cb62d to 6c583620ff7ad3235b135af7107c0543467eecfa, preserve the recorded exclusions/triggers, and link the next 98-source cycle in #497. Correct the documentation record to describe #483’s final compact-AGENTS resolution.

The source audit accounts for all 797 sources in the exact range, restores the omitted merged-port references, corrects port-SHA/source-SHA confusion in earlier PRs, and accounts for #481’s 18 pre-cursor gaps plus #461’s two Electron sources. No source is left unclassified and no Git ancestry baseline changes.

Validation: #483 passed fresh CI on 2863392ce000a5f81f1bdfa53efa43b43d057e99; #480 passed fresh CI on a6087ab9e4b0f7459f72b59ca843a5881915bddd. All five final/constituent PRs (#479–#483) are verified merged ancestors of ae27742ce14c112cc0c310e8aa5d780c5ad19205. Adversarial ledger review checked the exact range, retained partial exclusions and pre-cursor records, one toolchain row, and the next-cycle lower bound. The changed Markdown file passes the pinned formatter and git diff --check; no runtime code changes.

The automatic relay deployment on the #480 merge succeeded (applied): 3 resources planned for update, 16 no-ops, no replacements, and explicit [RelayNeonProject] updated, [RelayHyperdrive] updated, [Api] updated. The successful Neon update runs beta.76’s migration pipeline, which converts legacy relay_migrations bookkeeping in place while preserving applied history. The workflow emits no per-table conversion log, so the schema was not independently queried. No manual deployment was run.

Closes #414. Next cycle: #497, frozen at 4a4c6dd2adc350a68ba18bb28b24b5a7e4660dab, 98 sources / 571 unique paths / zero patch-equivalent sources; implementation remains to be reviewed there.

Model and harness: GPT-6 in 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 5:53am 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	14.1 KiB	14.0 KiB	−26 B (−0.2%)	15.1 KiB	✅
Codex	Thread snapshot wire	7.2 KiB	7.2 KiB	−1 B (−0.0%)	7.3 KiB	✅
Codex	Live turn WebSocket wire	6.9 KiB	6.8 KiB	−25 B (−0.4%)	7.8 KiB	✅
Codex	Live turn WebSocket decoded	58.8 KiB	58.8 KiB	−44 B (−0.1%)	66.4 KiB	✅
Codex	Live turn messages	10	9	−1 (−10.0%)	21	✅
Claude	Total thread wire	14.1 KiB	14.0 KiB	−21 B (−0.1%)	15.1 KiB	✅
Claude	Thread snapshot wire	7.2 KiB	7.2 KiB	+9 B (+0.1%)	7.3 KiB	✅
Claude	Live turn WebSocket wire	6.9 KiB	6.8 KiB	−30 B (−0.4%)	7.8 KiB	✅
Claude	Live turn WebSocket decoded	59.7 KiB	59.7 KiB	0 B (0.0%)	66.4 KiB	✅
Claude	Live turn messages	10	10	0 (0.0%)	21	✅

Baseline: ae27742 · PR result: fcb1318 · 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.