Adopt new Claude Code features: agent memory, context fork, worktree isolation, SubagentStop hook

- Add `memory: project` to 14 specialist agents for cross-session learning
- Add `context: fork` + `agent:` to 6 analysis skills to preserve main context
- Add `isolation: worktree` to prototyper agent for safe throwaway experiments
- Add SubagentStop hook to complete agent audit trail (start + stop logging)

Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>

.claude/agents/prototyper.md

@@ -4,6 +4,7 @@ description: "Rapid prototyping specialist for pre-production. Builds quick, thr
 tools: Read, Glob, Grep, Write, Edit, Bash
 model: sonnet
 maxTurns: 25
+isolation: worktree
 ---
 
 You are the Prototyper for an indie game project. Your job is to build things
@@ -60,6 +61,14 @@ Before writing any code:
 - Rules are your friend — when they flag issues, they're usually right
 - Tests prove it works — offer to write them proactively
 
+### Worktree Isolation
+
+This agent runs in `isolation: worktree` mode by default. All prototype code is
+written in a temporary git worktree — an isolated copy of the repository. If the
+prototype is killed or abandoned, the worktree is automatically cleaned up with
+no trace in the main working tree. If the prototype produces useful results, the
+worktree branch can be reviewed before merging.
+
 ### Core Philosophy: Speed Over Quality
 
 Prototype code is disposable. It exists to validate an idea as quickly as