diff --git a/CLAUDE.md b/CLAUDE.md
index 8f5e222..19ff454 100644
--- a/CLAUDE.md
+++ b/CLAUDE.md
@@ -137,9 +137,9 @@ and Fable 5 through the `claude` tool, Sol and Terra through `codex`.
 | Difficulty | Model | Spawn |
 |---|---|---|
 | `difficulty/hard` | Opus 5 | `aoe add <path> --tool claude --extra-args "--model claude-opus-5" -l` |
-| `difficulty/moderate` | Sol | `aoe add <path> --tool codex --extra-args "-m gpt-5.6-sol" -l` |
-| `difficulty/easy` | Terra | `aoe add <path> --tool codex --extra-args "-m gpt-5.6-terra" -l` |
-| `difficulty/trivial` | Fable 5 | `aoe add <path> --tool claude --extra-args "--model claude-fable-5" -l` |
+| `difficulty/moderate` | Fable 5 | `aoe add <path> --tool claude --extra-args "--model claude-fable-5" -l` |
+| `difficulty/easy` | Sol | `aoe add <path> --tool codex --extra-args "-m gpt-5.6-sol" -l` |
+| `difficulty/trivial` | Terra | `aoe add <path> --tool codex --extra-args "-m gpt-5.6-terra" -l` |
 
 Use `-w <branch> -b` to put each session in its own git worktree, which is what
 makes parallel issues safe.
@@ -149,8 +149,9 @@ unattended run:
 
 - `gpt-5.6-terra` is confirmed — it is the default in `~/.codex/config.toml`.
   `gpt-5.6-sol` is inferred from the sibling naming and unverified.
-- The Fable-5-at-trivial and Sol-above-Terra orderings are assumptions. If they
-  are the wrong way round, swap the rows; nothing else depends on the ordering.
+- Opus 5 and Fable 5 are both frontier models; the split between them at hard
+  and moderate is about reasoning depth, not capability tier. Sol above Terra is
+  an assumption — swap the rows if it is backwards.
 
 Escalate one tier if a session fails CI twice on the same issue. Never
 de-escalate mid-issue.