vault: align ECC docs with current repo and add orchestration manual

Updated ECC notes to match the current state of affaan-m/everything-claude-code
docs/token-optimization.md and docs/SKILL-PLACEMENT-POLICY.md:

- Drop CLAUDE_AUTOCOMPACT_PCT_OVERRIDE recommendations (now warned against
  upstream — variable can only lower threshold, opposite of intent)
- Add CLAUDE_CODE_SUBAGENT_MODEL=haiku as the new core token-saving setting
- Flag the default `memory` MCP for disablement (no skill/agent/hook references)
- Add Skill Placement Policy section (curated/learned/imported/evolved + provenance)
- Cover missing commands: /checkpoint, /sessions, /security-scan, /claw, /projects

Add new resource: ECC 编排实战手册.md (721 lines). Six orchestration patterns
(dmux+worktree, sequential claude -p, continuous-claude, Ralphinho RFC-DAG,
santa-loop, Task in-process) with real commands, real plan.json structures,
real CLI flags, and explicit "could not verify online" markers for /multi-*
and /feature-dev. All sourced to specific commands/*.md or skills/*.md files.

Cross-link the new manual from 完整指南 and 用法速查.

4 - Resources/Claude-Code/Everything Claude Code 完整指南.md

@@ -83,6 +83,44 @@ Legacy shim 仍然可用（向后兼容），只是内部转发到对应 skill
 
 Rules 新增：java, kotlin, dart, csharp, cpp, rust, perl, php, web, zh (中文)
 
+### Skill 放置策略 (新规范)
+
+仓库新增 `docs/SKILL-PLACEMENT-POLICY.md`，明确 4 类 skill 的放置位置和归属：
+
+| 类型 | 路径 | 提交仓库 | provenance |
+|------|------|---------|-----------|
+| **Curated**（人工策划） | `skills/` (repo) | ✅ | 用 frontmatter `origin` 字段 |
+| **Learned**（`/learn` 自动产出） | `~/.claude/skills/learned/` | ❌ | 必须 `.provenance.json` |
+| **Imported**（外部导入） | `~/.claude/skills/imported/` | ❌ | 必须 `.provenance.json` |
+| **Evolved**（instinct 聚类产出） | `~/.claude/homunculus/evolved/skills/` | ❌ | 继承自源 instinct |
+
+`.provenance.json` 必填字段：`source`、`created_at`、`confidence` (0-1)、`author`。
+
+**实操含义**：
+- `/learn` 学到的 skill 自动落到 `~/.claude/skills/learned/` —— 不要手动提交进 repo
+- `validate-skills.js` 只校验 `skills/` 下的 curated skill，忽略其他三类
+- 跨项目分享走 `/instinct-export` + `/instinct-import`，不直接 copy 文件
+
+详见 [docs/SKILL-PLACEMENT-POLICY.md](https://github.com/affaan-m/everything-claude-code/blob/main/docs/SKILL-PLACEMENT-POLICY.md)。
+
+### Token 优化新基线
+
+仓库 `docs/token-optimization.md` 当前推荐配置：
+
+```json
+{
+  "model": "sonnet",
+  "env": {
+    "MAX_THINKING_TOKENS": "10000",
+    "CLAUDE_CODE_SUBAGENT_MODEL": "haiku"
+  }
+}
+```
+
+- **`CLAUDE_CODE_SUBAGENT_MODEL=haiku`** 是新的省钱大头，子 agent 跑 Haiku 比默认便宜约 80%
+- ⚠️ **`CLAUDE_AUTOCOMPACT_PCT_OVERRIDE` 已被官方警告弃用** —— 该变量在新版 Claude Code 上只能让压缩更早触发，与延后目的相反，移除即可
+- `memory` MCP 默认装但无 skill/agent/hook 引用 → 项目级 `disabledMcpServers` 关掉
+
 ---
 
 ## 全部 65 Skills
@@ -266,6 +304,14 @@ Rules 新增：java, kotlin, dart, csharp, cpp, rust, perl, php, web, zh (中文
 ### v1.8.0 新增
 `/loop-start` `/loop-status` `/model-route` `/quality-gate` `/harness-audit` `/promote`
 
+### 验证 & 会话
+`/checkpoint`(verification-loop 检查点)`/verify`(完整验证管道)`/sessions`(会话历史)`/security-scan`(AgentShield 扫描)
+
+### REPL & 工具
+`/claw`(NanoClaw REPL，模型路由 + 技能热加载 + 会话分支)`/projects`(continuous-learning v2 项目管理)`/setup-pm`(包管理器配置)
+
+> 完整命令 → agent 映射见仓库 [docs/COMMAND-AGENT-MAP.md](https://github.com/affaan-m/everything-claude-code/blob/main/docs/COMMAND-AGENT-MAP.md)。
+
 ---
 
 ## Hooks 系统
@@ -303,6 +349,7 @@ ECC_DISABLED_HOOKS="pre:bash:tmux-reminder,post:edit:typecheck"
 ### Resources
 - [[Everything Claude Code 方法论与最佳实践]]
 - [[Everything Claude Code 用法速查]]
+- [[ECC 编排实战手册]] — 6 种编排模式真实命令 + 实战搭配 + 来源标注
 - [[Autonomous Loops 自主循环模式]]
 - [[dmux 多Agent并行编排]]
 - [[Ralphinho RFC-DAG 编排模式]]