diff --git a/.github/agents/pr/PLAN-TEMPLATE.md b/.github/agents/pr/PLAN-TEMPLATE.md
index 11901d56dd91..6e9836d7d9fd 100644
--- a/.github/agents/pr/PLAN-TEMPLATE.md
+++ b/.github/agents/pr/PLAN-TEMPLATE.md
@@ -56,10 +56,10 @@ See `SHARED-RULES.md` for complete details. Key points:
 *(Only if Gate ✅ PASSED)*
 
 **Round 1: Run try-fix with each model (SEQUENTIAL)**
-- [ ] claude-sonnet-4.5
+- [ ] claude-sonnet-4.6
 - [ ] claude-opus-4.6
 - [ ] gpt-5.2
-- [ ] gpt-5.2-codex
+- [ ] gpt-5.3-codex
 - [ ] gemini-3-pro-preview
 - [ ] ⛔ If blocker: STOP, report, ask user
 - [ ] Record: approach, result, files, failure analysis
diff --git a/.github/agents/pr/SHARED-RULES.md b/.github/agents/pr/SHARED-RULES.md
index 70662faccd9b..1347d92e659a 100644
--- a/.github/agents/pr/SHARED-RULES.md
+++ b/.github/agents/pr/SHARED-RULES.md
@@ -131,10 +131,10 @@ Phase 4 uses these 5 AI models for try-fix exploration (run SEQUENTIALLY):
 
 | Order | Model |
 |-------|-------|
-| 1 | `claude-sonnet-4.5` |
+| 1 | `claude-sonnet-4.6` |
 | 2 | `claude-opus-4.6` |
 | 3 | `gpt-5.2` |
-| 4 | `gpt-5.2-codex` |
+| 4 | `gpt-5.3-codex` |
 | 5 | `gemini-3-pro-preview` |
 
 **Note:** The `model` parameter is passed to the `task` tool, which supports model selection. This is separate from agent YAML frontmatter (which is VS Code-only).
diff --git a/.github/agents/pr/post-gate.md b/.github/agents/pr/post-gate.md
index f030801ee257..7d80660a0751 100644
--- a/.github/agents/pr/post-gate.md
+++ b/.github/agents/pr/post-gate.md
@@ -202,10 +202,10 @@ Update the state file:
   ```
   | Model | Round 2 Response |
   |-------|------------------|
-  | claude-sonnet-4.5 | NO NEW IDEAS |
+  | claude-sonnet-4.6 | NO NEW IDEAS |
   | claude-opus-4.6 | NO NEW IDEAS |
   | gpt-5.2 | NO NEW IDEAS |
-  | gpt-5.2-codex | NO NEW IDEAS |
+  | gpt-5.3-codex | NO NEW IDEAS |
   | gemini-3-pro-preview | NO NEW IDEAS |
   ```
 - [ ] Fix Candidates table has numbered rows for each try-fix attempt