amadan
docs: Fable joins the delegation ladder — outcome-changing work only
Paul's ruling 2026-08-03: use Fable where it's appropriate — the tier that changes the outcome, not the speed. Converge-grade implementation, whole-branch reviews, fresh-context deep work. Everything below it stays exactly as tiered; today's evidence is that Sonnet-with-a-precise-brief and Opus reviews keep catching what they are meant to catch.
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
1 file changed,
+6
−0
CLAUDE.md+6 −0
diff --git a/CLAUDE.md b/CLAUDE.md| index b1797c72e..57f6a6c71 100644 |
| --- a/CLAUDE.md |
| +++ b/CLAUDE.md |
| @@ -15,6 +15,12 @@ have Opus review anything that grants authority or touches production paths: |
| interfaces, established conventions to follow. |
| - **Opus** — design judgment, multi-file integration, anything security- or |
| production-touching; and the **reviewer** for work in those areas. |
| +- **Fable** — where the tier changes the outcome, not the speed (Paul, |
| + 2026-08-03): implementation whose correctness rests on several interacting |
| + invariants at once (converge-grade, root-side, fleet-wide blast radius), |
| + whole-branch final reviews, and deep work that wants a fresh |
| + full-capability context window. It burns usage budget fastest — never |
| + spend it where a precise Sonnet brief or an Opus review already fits. |
| Always name the model explicitly when dispatching; an omitted model silently |
| inherits the session's (usually the most expensive). A precise brief beats a |