refactor: wave 6 — ratchet mypy coverage, link clean-code standard (#53)

Polish pass closing out the refactor series.

pyproject.toml
--------------
- Drop skillforge/engine/scorer.py from the mypy exclusion list. After
  Wave 2's logging cleanup it now type-checks clean at the default
  level. Mypy now covers 65 source files (up from 64).

CLAUDE.md
---------
- Add docs/clean-code.md to the Key Reference Documents list.
- Expand the Code Style TL;DR with the five load-bearing rules the
  refactor waves actually enforced (no bare except, no print in
  library code, no mutable module globals, no raw fetch in React
  components, file-size ceilings) so the standard is discoverable
  from the project root.

QA
--
  ruff check skillforge     - clean
  mypy skillforge           - 65 files pass
  pytest tests/             - 403 passed, 2 skipped
  frontend build/lint/test  - all green

Co-authored-by: Matt (via Claude Code) <matt@skillforge.local>
Co-authored-by: Claude Opus 4.7 (1M context) <noreply@anthropic.com>

Author: 3 people

CLAUDE.md

@@ -54,6 +54,7 @@ Core loop: `skillforge/engine/evolution.py` (molecular) + `skillforge/engine/var
 
 ## Key Reference Documents
 - `docs/how-skld-works.md` — **start here**: full system overview for first-time readers.
+- `docs/clean-code.md` — the review rubric every PR is held to (naming, functions, errors, data, async, functional idioms, React/TS, testing, comments, checklist).
 - `plans/SPEC-V2.0.md` — v2.0 architecture spec (taxonomy, agents, variants, evaluation, data model).
 - `plans/PLAN-V2.0.md` — v2.0 implementation plan (5 phases, 15 waves, file-by-file).
 - `plans/SPEC-V2.1.md` — v2.1 architecture spec (controlled evaluation environments, tiered challenge pools, train/test separation).
@@ -181,11 +182,17 @@ When running autonomously (overnight work, no active user):
 - **On hard block**: document in Progress Tracker with `[BLOCKED: reason]` marker, write a journal entry explaining what was tried, skip to any independent work, stop cleanly if no independent work remains.
 
 ## Code Style
+Full rubric: **`docs/clean-code.md`**. TL;DR:
 - Type hints everywhere. Dataclasses for internal models, Pydantic only at API boundary.
 - Async throughout — evolution engine is fully async.
 - No classes where functions suffice.
-- Short functions, clear names, minimal comments.
+- Short functions (≤50 lines), clear names, minimal comments.
 - Prefer composition over inheritance.
+- No bare `except Exception` — typed exceptions from `skillforge/errors.py`.
+- No `print` in library code — `logging` module with `logger.exception` inside excepts.
+- No mutable module globals — use `skillforge/engine/run_registry.py` or similar injected state.
+- No raw `fetch()` in React components — typed hooks from `frontend/src/api/hooks/`.
+- File ceilings: 500 LOC (Python), 400 LOC (TSX). Split into packages when they grow.
 
 ## Testing
 - Unit tests mock the Agent SDK.

pyproject.toml

@@ -89,7 +89,6 @@ exclude = [
     "^skillforge/engine/evolution\\.py$",
     "^skillforge/engine/export\\.py$",
     "^skillforge/engine/sandbox\\.py$",
-    "^skillforge/engine/scorer\\.py$",
     "^skillforge/main\\.py$",
     "^skillforge/models/run\\.py$",
 ]