larkway 0.3.7 → 0.3.9

package/README.md

@@ -8,7 +8,7 @@
 
 You @ the bot in a Feishu thread. It runs on your machine — reading your real codebase, executing commands, opening MRs — and posts the result back. You define what the agent knows and what it can do. Larkway just carries the messages.
 
-**Current release: v0.3.6**
+**Current release: v0.3.9**
 
 ---
 
@@ -113,7 +113,7 @@ Larkway never injects `ANTHROPIC_API_KEY` or any other API key. The subprocess i
 | **L2 identity memory** | "Who I am, what I must not do, where to find the workflow" (thin) | `~/.larkway/bots/<id>.memory.md` |
 | **L3 workflow** | State machine, gates, commands — the actual job | **Your business repo**: `AGENTS.md`, `CLAUDE.md`, `.agents/skills/`, `.claude/skills/` |
 
-Secrets live only in `~/.larkway/.env` (mode 0600). Config and memory contain no secrets and are safe to share via the central config library.
+Secrets live only in `~/.larkway/.env` (mode 0600). Config and memory contain no secrets.
 
 ---
 
@@ -121,7 +121,6 @@ Secrets live only in `~/.larkway/.env` (mode 0600). Config and memory contain no
 
 - **Multiple bots on one bridge** — a read-only Q&A bot and a write-capable engineering bot can share the same process, each with its own L1/L2/L3 definition
 - **Web UI** — `larkway ui` opens a local management dashboard (127.0.0.1 + token); create bots, edit memory, watch live logs
-- **Central config library** — `larkway promote <id>` pushes bot config (no secrets) to a shared git repository so teammates can `larkway sync` and discover bots; the library stores config, not deployment state
 - **Session continuity** — every Feishu thread maps to a persistent `session_id`; the agent remembers what it did in prior turns
 - **Agent Workspace** — per-thread git worktrees; the agent can run multiple threads concurrently without git conflicts
 - **Codex runtime pre-checks** — `larkway doctor` validates Codex state directory writability before start
@@ -150,9 +149,6 @@ Secrets live only in `~/.larkway/.env` (mode 0600). Config and memory contain no
 | `larkway memory edit <id>` | Edit L2 identity memory |
 | `larkway perms <id>` | Adjust L1 permissions |
 | `larkway ui` | Start local web management UI |
-| `larkway central set \| show \| unset` | Connect/disconnect central config library |
-| `larkway promote <id>` | Push bot config (no secrets) to central library |
-| `larkway sync` | Pull bot configs from central library |
 | `larkway update` | Upgrade Larkway and restart bridge |
 
 ---

package/README.zh.md

@@ -8,7 +8,7 @@
 
 你在飞书话题里 @ bot，它在你的机器上运行——读真实代码库、执行命令、开 MR——把结果贴回飞书。你定义 agent 知道什么、能做什么。Larkway 只负责传递消息。
 
-**当前版本：v0.3.6**
+**当前版本：v0.3.9**
 
 ---
 
@@ -113,7 +113,7 @@ Larkway 不注入 `ANTHROPIC_API_KEY` 或任何其他 API key。子进程继承
 | **L2 身份 memory** | "我是谁、禁止什么、工作流指针"（薄） | `~/.larkway/bots/<id>.memory.md` |
 | **L3 工作流** | 状态机、gate、命令——实际的工作内容 | **业务 repo**：`AGENTS.md`、`CLAUDE.md`、`.agents/skills/`、`.claude/skills/` |
 
-密钥只存在本机 `~/.larkway/.env`（权限 0600）。配置和 memory 不含密钥，可以安全地通过中心配置库共享。
+密钥只存在本机 `~/.larkway/.env`（权限 0600）。配置和 memory 不含密钥。
 
 ---
 
@@ -121,7 +121,6 @@ Larkway 不注入 `ANTHROPIC_API_KEY` 或任何其他 API key。子进程继承
 
 - **一个 bridge 跑多个 bot** —— 只读答疑 bot 和有写权限的工程 bot 可以共用同一个进程，各自有独立的 L1/L2/L3 定义
 - **Web 管理面** —— `larkway ui` 打开本地管理后台（127.0.0.1 + token），可以建 bot、编辑 memory、查看实时日志
-- **中心配置库** —— `larkway promote <id>` 把 bot 配置（不含密钥）推送到共享 git 仓库，队友可以 `larkway sync` 发现和复用；库只存配置，不存部署状态
 - **Session 续接** —— 每个飞书话题映射到持久 `session_id`，agent 记得之前做了什么
 - **Agent Workspace** —— 每个话题独立 git worktree，agent 并发处理多个话题不冲突
 - **Codex 运行时预检** —— `larkway doctor` 在启动前验证 Codex 状态目录可写
@@ -150,9 +149,6 @@ Larkway 不注入 `ANTHROPIC_API_KEY` 或任何其他 API key。子进程继承
 | `larkway memory edit <id>` | 编辑 L2 身份 memory |
 | `larkway perms <id>` | 调整 L1 权限 |
 | `larkway ui` | 启动本地 Web 管理面 |
-| `larkway central set \| show \| unset` | 连接/断开中心配置库 |
-| `larkway promote <id>` | 把 bot 配置（不含密钥）推送到中心库 |
-| `larkway sync` | 从中心库拉 bot 配置 |
 | `larkway update` | 升级 Larkway 并重启 bridge |
 
 ---