wangchuan 6.0.3 → 7.0.0

package/skill/SKILL.md

@@ -1,11 +1,11 @@
 ---
 name: wangchuan
-version: 2.0.0
+version: 3.0.0
 description: >-
   Encrypt and sync AI agent configs, memories, skills, and MCP servers across environments via a private Git repo.
   Supports Claude, OpenClaw, Gemini, CodeBuddy, WorkBuddy, Cursor, and Codex.
-  TRIGGER when: user mentions syncing AI memories/configs, wangchuan/忘川, cross-machine agent setup, backup/restore agent settings, memory sync, skill distribution, MCP server sync, agent migration, master key export, sync status, watch daemon, workspace leakage, environment management, snapshot, pushing/pulling memories, conflict resolution, syncing between agents, rolling back/restoring versions, switching environments, health check, customizing agent paths, restore cloud memories, or any commands: init, restore, sync, status, watch, doctor, memory, env, snapshot, lang.
-  Triggers: 忘川、wangchuan、同步记忆、同步配置、同步技能、同步MCP、初始化忘川、备份记忆、恢复记忆、恢复云端记忆、绑定云端记忆、迁移agent、导出密钥、轮换密钥、同步状态、健康检查、多环境、快照、跨机器同步、agent记忆、配置路径、查看技能、删除技能、新增技能、修改技能、自定义agent、MCP服务器、新增MCP、删除MCP、修改MCP、查看MCP、写记忆、删除记忆、修改记忆、查看记忆、广播记忆、复制记忆、推送记忆、拉取记忆、同步到agent、冲突、合并记忆、回退记忆、回滚、恢复版本、切换环境、忘川状态、新建环境、删除环境、查看环境、切换语言、升级忘川、更新忘川、后台同步、启动watch、环境泄漏、旧环境文件、sync memories、push memory、pull memory、rollback、switch environment、custom agent、MCP server、health status、rotate key、switch language、upgrade wangchuan、update wangchuan、restore memory、restore cloud.
+  TRIGGER when: user mentions syncing AI memories/configs, wangchuan/忘川, cross-machine agent setup, backup/restore agent settings, memory sync, skill distribution, MCP server sync, agent migration, master key export, sync status, workspace leakage, environment management, snapshot, pushing/pulling memories, conflict resolution, syncing between agents, rolling back/restoring versions, switching environments, health check, customizing agent paths, restore cloud memories, or any commands: init, restore, pull, push, status, doctor, memory, env, snapshot, lang.
+  Triggers: 忘川、wangchuan、同步记忆、同步配置、同步技能、同步MCP、初始化忘川、备份记忆、恢复记忆、恢复云端记忆、绑定云端记忆、迁移agent、导出密钥、轮换密钥、同步状态、健康检查、多环境、快照、跨机器同步、agent记忆、配置路径、查看技能、删除技能、新增技能、修改技能、自定义agent、MCP服务器、新增MCP、删除MCP、修改MCP、查看MCP、写记忆、删除记忆、修改记忆、查看记忆、广播记忆、复制记忆、推送记忆、拉取记忆、同步到agent、冲突、合并记忆、回退记忆、回滚、恢复版本、切换环境、忘川状态、新建环境、删除环境、查看环境、切换语言、升级忘川、更新忘川、环境泄漏、旧环境文件、sync memories、push memory、pull memory、rollback、switch environment、custom agent、MCP server、health status、rotate key、switch language、upgrade wangchuan、update wangchuan、restore memory、restore cloud.
   DO NOT TRIGGER when: general git operations, unrelated CLI tools, or AI model APIs.
 ---
 
@@ -23,16 +23,16 @@ If `~/.wangchuan/config.json` does not exist → Read [references/install-setup.
 ```
 wangchuan init     [--repo <url>] [--key <path>]           One-time setup
 wangchuan restore  --repo <url> --key <key>                Restore cloud to new machine
-wangchuan sync     [-a <agent>] [-n] [-o <pat>] [-x <pat>] Smart bidirectional sync
+wangchuan pull     [-a <agent>] [-o <pat>] [-x <pat>]      Pull cloud data to local
+wangchuan push     [-a <agent>] [-n] [-y] [-o <pat>] [-x <pat>] Push local changes to cloud
 wangchuan status   [-v]                                     Health + diff summary
-wangchuan watch    [-i <min>]                               Pull-only background daemon
 wangchuan doctor   [--key-export|--key-rotate|--setup]      Diagnose + auto-fix
 wangchuan memory   list|show|copy|broadcast [-a <agent>]    Memory management
 wangchuan env      list|create|switch|current|delete        Multi-environment
 wangchuan snapshot save|list|restore|delete [name]          Snapshots
 wangchuan lang     [zh|en]                                  Display language
 ```
-Aliases: `sync`→`s`, `status`→`st`, `snapshot`→`snap`. All support `--agent` / `-a` filter.
+Aliases: `push`→`s`, `status`→`st`, `snapshot`→`snap`. All support `--agent` / `-a` filter.
 
 ## Task Routing — Load the right reference BEFORE acting
 
@@ -54,7 +54,7 @@ Aliases: `sync`→`s`, `status`→`st`, `snapshot`→`snap`. All support `--agen
 |---------|-----------|----------|
 | `wangchuan init` | Interactive prompt fails | Must pass `--repo <url>` |
 | `wangchuan restore` | Interactive prompt fails | Must pass `--repo <url>` and `--key <key>` |
-| `wangchuan sync` | Pending confirmations skipped | Must pass `-y` |
+| `wangchuan push` | Pending confirmations skipped | Must pass `-y` |
 | `wangchuan env create` | Memory import prompt fails | Auto-forks (OK) |
 
 ## Customizing Agent Workspace Paths
@@ -63,52 +63,41 @@ Default paths are in `~/.wangchuan/config.json`. To change (e.g. Claude to `~/.c
 ```bash
 jq '.profiles.default.claude.workspacePath = "~/.claude-internal"' ~/.wangchuan/config.json > /tmp/wc-cfg.json && mv /tmp/wc-cfg.json ~/.wangchuan/config.json
 ```
-Tell user: "Path updated locally. Run `wangchuan sync` to push to cloud." If user confirms: `wangchuan sync -y`
+Tell user: "Path updated locally. Run `wangchuan push` to push to cloud." If user confirms: `wangchuan push -y`
 
 ## Sync Protocol — Rules agents MUST follow
 
 ### On session start
 ```bash
-wangchuan status
-wangchuan sync -n   # dry-run preview of pending changes
+wangchuan pull
 ```
-If there are pending changes, show the preview to the user and ask: "There are pending changes. Want to sync now?" If yes → `wangchuan sync -y`.
-
-### After EVERY skill invocation (mandatory)
-After completing any user instruction, check if the watch daemon is running and start it if not:
-```bash
-# Check if watch is running, start if not (pull-only daemon)
-pgrep -f 'wangchuan.*watch' >/dev/null 2>&1 || nohup wangchuan watch >/dev/null 2>&1 &
-```
-This ensures cloud memories are always pulled in the background. Watch is pull-only — it never pushes.
+Pull cloud data to local on session start. No push needed.
 
 ### Rule summary
 
 | Event | Action |
 |-------|--------|
-| Created/modified a **shared** skill or custom agent | Copy to all other agents → tell user: "Changes saved locally. Run `wangchuan sync` to push to cloud." → if user confirms: `wangchuan sync -y` |
-| Created/modified a **new/local** skill or custom agent | Resource stays in current agent by default. Only distribute if user **explicitly asks** to share → ask which agents → copy → register in shared-registry → tell user: "Changes saved locally. Run `wangchuan sync` to push to cloud." → if user confirms: `wangchuan sync -y` |
-| **Deleted** a skill or custom agent | **Always ask** → inform shared status → remove selected → tell user: "Changes saved locally. Run `wangchuan sync` to push to cloud." → if user confirms: `wangchuan sync -y` |
-| Created/modified an **MCP server** | Resource stays in current agent by default. Each agent's MCP config is independent. Only copy to other agents if user **explicitly asks** → ask which agents → jq write → tell user: "Changes saved locally. Run `wangchuan sync` to push to cloud." → if user confirms: `wangchuan sync -y` |
-| **Deleted** an MCP server | Ask user which agents → jq del → tell user: "Changes saved locally. Run `wangchuan sync` to push to cloud." → if user confirms: `wangchuan sync -y` |
-| Created/modified **memory** | Ask user → broadcast/copy if yes → tell user: "Changes saved locally. Run `wangchuan sync` to push to cloud." → if user confirms: `wangchuan sync -y` |
-| **Deleted** memory | Ask user which agents → rm → tell user: "Changes saved locally. Run `wangchuan sync` to push to cloud." → if user confirms: `wangchuan sync -y` |
+| Created/modified a **shared** skill or custom agent | Copy to all other agents → tell user: "Changes saved locally. Run `wangchuan push` to push to cloud." → if user confirms: `wangchuan push -y` |
+| Created/modified a **new/local** skill or custom agent | Resource stays in current agent by default. Only distribute if user **explicitly asks** to share → ask which agents → copy → register in shared-registry → tell user: "Changes saved locally. Run `wangchuan push` to push to cloud." → if user confirms: `wangchuan push -y` |
+| **Deleted** a skill or custom agent | **Always ask** → inform shared status → remove selected → tell user: "Changes saved locally. Run `wangchuan push` to push to cloud." → if user confirms: `wangchuan push -y` |
+| Created/modified an **MCP server** | Resource stays in current agent by default. Each agent's MCP config is independent. Only copy to other agents if user **explicitly asks** → ask which agents → jq write → tell user: "Changes saved locally. Run `wangchuan push` to push to cloud." → if user confirms: `wangchuan push -y` |
+| **Deleted** an MCP server | Ask user which agents → jq del → tell user: "Changes saved locally. Run `wangchuan push` to push to cloud." → if user confirms: `wangchuan push -y` |
+| Created/modified **memory** | Ask user → broadcast/copy if yes → tell user: "Changes saved locally. Run `wangchuan push` to push to cloud." → if user confirms: `wangchuan push -y` |
+| **Deleted** memory | Ask user which agents → rm → tell user: "Changes saved locally. Run `wangchuan push` to push to cloud." → if user confirms: `wangchuan push -y` |
 | **Files deleted from cloud** detected on pull | Automatically deleted from local workspace (cloud is source of truth). All changes preserved in git history for rollback. |
-| Updated other config | Tell user: "Changes saved locally. Run `wangchuan sync` to push to cloud." → if user confirms: `wangchuan sync -y` |
-| **Any skill invocation completes** | **Ensure watch daemon is running** (see above) |
+| Updated other config | Tell user: "Changes saved locally. Run `wangchuan push` to push to cloud." → if user confirms: `wangchuan push -y` |
 
 **IMPORTANT: Resources are agent-specific by default.** Skills, custom agents, and MCP servers are NOT auto-distributed or auto-merged across agents. They stay in the agent where they were created. Cross-agent sharing only happens when the user explicitly requests it.
 
-**IMPORTANT: Pushing to cloud NEVER happens automatically.** After any CRUD operation, inform the user that changes are saved locally and ask if they want to sync to cloud. Only run `wangchuan sync -y` when the user explicitly confirms.
+**IMPORTANT: Pushing to cloud NEVER happens automatically.** After any CRUD operation, inform the user that changes are saved locally and ask if they want to push to cloud. Only run `wangchuan push -y` when the user explicitly confirms.
 
 ### Environment awareness
 
-All push/pull/watch operations target the **current environment's branch only**. Key rules:
+All push/pull operations target the **current environment's branch only**. Key rules:
 - **Push/pull**: always bound to `cfg.environment` → the current env's git branch
-- **Watch**: pulls from current env only. After `env switch`, restart watch
-- **Cross-env access**: NOT supported — must `env switch` first, then sync
+- **Cross-env access**: NOT supported — must `env switch` first, then pull/push
 - **Workspace leakage**: switching env does NOT delete local files from the previous env. After switch, stale files from old env may remain in `~/.claude/skills/` etc. Do NOT push these to the new env — run `wangchuan status -v` to check `localOnly` files first
-- **Pull from another env**: if user asks "pull work env's memory", must switch first: `wangchuan env switch work` → auto-syncs
+- **Pull from another env**: if user asks "pull work env's memory", must switch first: `wangchuan env switch work` → auto-pulls
 
 ### --agent filter values
 

package/skill/references/environment.md

@@ -10,14 +10,13 @@ Environments map to git branches: `default` → `main`, `<name>` → `env/<name>
 
 | Operation | Environment behavior |
 |-----------|---------------------|
-| `sync` (push/pull) | Always targets current env's branch. Cannot push to or pull from another env. |
-| `watch` | Pulls from current env only. After `env switch`, **must restart watch**. |
-| Skill/agent/MCP CRUD | Changes apply to local workspace, pushed to current env's branch on sync. |
-| `memory copy/broadcast` | Operates on local workspace files (shared). Pushed to current env on sync. |
+| `push` / `pull` | Always targets current env's branch. Cannot push to or pull from another env. |
+| Skill/agent/MCP CRUD | Changes apply to local workspace, pushed to current env's branch on push. |
+| `memory copy/broadcast` | Operates on local workspace files (shared). Pushed to current env on push. |
 | After `env switch` | Old env's local files may linger. Check `wangchuan status -v` for `localOnly` files before pushing — do NOT blindly push stale files to the new env. |
 | Pull from another env | Must `env switch <target>` first. No `--from-env` flag exists. |
 
-## Listing environments (指令 26)
+## Listing environments
 
 When user says "查看忘川环境列表" / "list environments":
 
@@ -45,7 +44,7 @@ git log origin/main --oneline -20 --format="%s" | grep -oP '\[([^\]]+)\]$' | sor
 
 For detailed health of a specific environment, switch to it and run `wangchuan status -v` (see inspect-status.md).
 
-## Current environment (指令 27)
+## Current environment
 
 When user says "看下忘川当前环境" / "which environment am I in":
 
@@ -56,7 +55,7 @@ wangchuan status -v
 
 Report: environment name, active machines count, health score, unsynced files, anomalies.
 
-## Creating a new environment (指令 28)
+## Creating a new environment
 
 When user says "新建忘川 xxx 环境" / "create xxx environment":
 
@@ -65,7 +64,7 @@ When user says "新建忘川 xxx 环境" / "create xxx environment":
 wangchuan env list
 ```
 If the name already exists, inform the user and ask: "Environment 'xxx' already exists. Would you like to: (1) Import current env's memories into it, or (2) Just switch to it?"
-- Import: `wangchuan env switch <name>` → then ask user if they want to sync: `wangchuan sync -y`
+- Import: `wangchuan env switch <name>` → then ask user if they want to push: `wangchuan push -y`
 - Switch only: `wangchuan env switch <name>`
 
 **Step 2: Ask about data initialization.**
@@ -92,7 +91,7 @@ wangchuan env switch <name>
 
 `env create` in non-TTY mode auto-forks with memories (no interactive prompt needed). To create an empty environment, there is no flag — the fork behavior is the default. If user explicitly wants empty, agent would need to create the branch manually.
 
-## Deleting an environment (指令 29)
+## Deleting an environment
 
 When user says "删除忘川 xxx 环境" / "delete xxx environment":
 
@@ -118,14 +117,14 @@ When user says "切换到 xxx 环境" / "switch to work environment":
 ```bash
 wangchuan status
 ```
-- **Few changes (≤3 files)**: warn briefly, ask user: "Push changes before switching?" If yes: `wangchuan sync -y`
+- **Few changes (≤3 files)**: warn briefly, ask user: "Push changes before switching?" If yes: `wangchuan push -y`
 - **Many changes or conflicts**: `wangchuan status -v` → show diff → ask "Push current changes first, or discard and switch?"
 
 **Step 2: Switch.**
 ```bash
 wangchuan env switch <name>
 ```
-Auto-switches branch, updates config, runs sync to pull target env's data.
+Auto-switches branch, updates config, runs pull to get target env's data.
 
 **Step 3: Post-switch checks.**
 
@@ -144,13 +143,7 @@ Warn user: "These files exist locally but not in the new environment. Do NOT pus
 
 **Note**: Since v1.9.0, pull automatically deletes local files that were removed from cloud (cloud is source of truth). All changes are preserved in git history for rollback.
 
-**Step 4: Restart watch daemon** (watch only pulls from current env's branch):
-```bash
-pkill -f 'wangchuan.*watch' 2>/dev/null; nohup wangchuan watch >/dev/null 2>&1 &
-```
-Watch mode **only pulls** cloud changes — it does NOT push local changes. Users must run `wangchuan sync` manually to push. If watch encounters a conflict it cannot auto-merge, it records it to `~/.wangchuan/pending-conflicts.json` — the next `wangchuan sync` will display these conflicts for user resolution.
-
-**Step 5: If user asked "pull memory from env X" without switching:**
+**Step 4: If user asked "pull memory from env X" without switching:**
 Explain that cross-env pull is not supported — must switch first:
 ```bash
 wangchuan env switch <target-env>   # switch → auto-pulls target env's data
@@ -192,7 +185,7 @@ Confirm with user before deleting.
 When user says "回退记忆" / "rollback" / "restore a previous version":
 
 **Step 1: Identify target.** Ask user's intent:
-- **Undo last sync** → find auto-snapshot
+- **Undo last push** → find auto-snapshot
 - **Specific time** → `wangchuan snapshot list`
 - **Recover deleted file** → `cd ~/.wangchuan/repo && git log --oneline --name-status -20`
 - **Revert specific change** → `git log` + `git show <hash> --stat`
@@ -213,20 +206,20 @@ cd ~/.wangchuan/repo && git show <hash>:<file-path>
 
 **Step 3: Execute.**
 
-Snapshot restore (restores locally, ask user to sync):
+Snapshot restore (restores locally, ask user to push):
 ```bash
 wangchuan snapshot restore <name>
 ```
-Tell user: "Snapshot restored locally. Run `wangchuan sync` to push to cloud." If user confirms: `wangchuan sync -y`
+Tell user: "Snapshot restored locally. Run `wangchuan push` to push to cloud." If user confirms: `wangchuan push -y`
 
-Single file from git (needs sync after):
+Single file from git (needs push after):
 ```bash
 cd ~/.wangchuan/repo && git checkout <hash> -- <file-path>
 ```
-Tell user: "File restored locally. Run `wangchuan sync` to push to cloud." If user confirms: `wangchuan sync -y`
+Tell user: "File restored locally. Run `wangchuan push` to push to cloud." If user confirms: `wangchuan push -y`
 
-Undo last sync:
+Undo last push:
 ```bash
-wangchuan snapshot list  # find pre-sync auto-snapshot (second most recent)
-wangchuan snapshot restore <pre-sync-snapshot>
+wangchuan snapshot list  # find pre-push auto-snapshot (second most recent)
+wangchuan snapshot restore <pre-push-snapshot>
 ```

package/skill/references/inspect-status.md

@@ -81,10 +81,10 @@ All of compact, plus:
 
 | Symptom | Meaning | Fix |
 |---------|---------|-----|
-| Freshness low | Haven't synced recently | `wangchuan sync -y` |
-| Coverage low | Local files missing | `wangchuan sync -y` (pull restores) |
+| Freshness low | Haven't synced recently | `wangchuan push -y` |
+| Coverage low | Local files missing | `wangchuan pull` (pull restores) |
 | Integrity low | Checksums mismatch | `wangchuan doctor` |
 | Encryption low | Sensitive files unencrypted | Review config profiles |
-| Conflict warning | Local+remote both changed | `wangchuan sync -y` → resolve markers |
+| Conflict warning | Local+remote both changed | `wangchuan push -y` → resolve markers |
 | Stale sync lock | Previous sync crashed | `wangchuan doctor` |
-| Pending distributions | Unprocessed sharing | `wangchuan sync -y` |
+| Pending distributions | Unprocessed sharing | `wangchuan push -y` |

package/skill/references/install-setup.md

@@ -26,44 +26,41 @@ npm update -g wangchuan
 wangchuan --version
 ```
 
-**Step 3: Pull cloud first, then sync local changes when ready.**
+**Step 3: Pull cloud first, then push local changes when ready.**
 
-After upgrade, the new version may include new/changed agent sync profiles (new syncDirs, syncFiles, jsonFields). The built-in `reconcileProfiles` mechanism automatically detects these changes when `sync` runs — it compares the built-in agent definitions against `config.json`, adds any missing entries, and saves the updated config. No manual config editing needed.
+After upgrade, the new version may include new/changed agent sync profiles (new syncDirs, syncFiles, jsonFields). The built-in `reconcileProfiles` mechanism automatically detects these changes when `push` runs — it compares the built-in agent definitions against `config.json`, adds any missing entries, and saves the updated config. No manual config editing needed.
 
-Tell the user: "Upgrade complete. Run `wangchuan sync` to pull cloud data and push any newly-discovered local files." When the user confirms:
+Tell the user: "Upgrade complete. Run `wangchuan pull` to get cloud data, then `wangchuan push` to push any newly-discovered local files." When the user confirms:
 
 ```bash
-# Pull cloud data and push local changes
-wangchuan sync -y
+# Pull cloud data
+wangchuan pull
+# Push local changes
+wangchuan push -y
 ```
 
-This single command:
+This flow:
 1. Auto-runs `reconcileProfiles` → detects new sync entries → updates `config.json`
 2. **Pulls latest from cloud first** (current env branch) — ensures no cloud data is lost
 3. Then pushes any newly-synced local files to cloud
 
-If the first sync shows "no changes" but you expect new files to be discovered, run sync again — the first run updates config.json, the second picks up newly-discovered files:
+If the first push shows "no changes" but you expect new files to be discovered, run push again — the first run updates config.json, the second picks up newly-discovered files:
 ```bash
-wangchuan sync -y   # second sync if needed
+wangchuan push -y   # second push if needed
 ```
 
-**Note**: Sync is NOT automatic after upgrade. The user must explicitly request it.
+**Note**: Push is NOT automatic after upgrade. The user must explicitly request it.
 
 **Step 4: Report results.**
 Tell the user:
 - New version number
-- Any new sync entries added (visible in sync output as newly-discovered files)
+- Any new sync entries added (visible in push output as newly-discovered files)
 - Files synced (pulled/pushed)
 - Current environment name
 
-**Step 5: Ensure watch daemon running.**
-```bash
-pgrep -f 'wangchuan.*watch' >/dev/null 2>&1 || nohup wangchuan watch >/dev/null 2>&1 &
-```
-
 **Complete upgrade flow:**
 ```
-npm update -g wangchuan → wangchuan --version → ask user to sync → wangchuan sync -y (if confirmed) → report → ensure watch
+npm update -g wangchuan → wangchuan --version → ask user to pull/push → wangchuan pull + wangchuan push -y (if confirmed) → report
 ```
 
 ## Initialization (when `~/.wangchuan/config.json` does not exist)
@@ -76,7 +73,7 @@ npm update -g wangchuan → wangchuan --version → ask user to sync → wangchu
 2. Run: `wangchuan init --repo <url>`
 3. Auto: generates key → clones → detects agents → extracts shared resources → pulls cloud data
 4. **Remind user to back up key**: `wangchuan doctor --key-export`
-5. After init, tell user: "Initialization complete. Run `wangchuan sync` when ready to push local data to cloud."
+5. After init, tell user: "Initialization complete. Run `wangchuan push` when ready to push local data to cloud."
 
 ## Restore (New Machine)
 
@@ -125,11 +122,6 @@ If only `default` exists → no need to ask, already synced.
 wangchuan status -v
 ```
 
-**Step 6: Ensure watch daemon.**
-```bash
-pgrep -f 'wangchuan.*watch' >/dev/null 2>&1 || nohup wangchuan watch >/dev/null 2>&1 &
-```
-
 ## Key management
 
 ```bash
@@ -147,7 +139,7 @@ wangchuan doctor --setup
 
 ## Key mismatch handling
 
-If `Key mismatch!` appears during sync:
+If `Key mismatch!` appears during push/pull:
 1. `wangchuan doctor --key-export` on the machine that last pushed
 2. Copy key hex to current machine
 3. `wangchuan init --key <hex>` or write to `~/.wangchuan/master.key`
@@ -161,8 +153,8 @@ cp -r wangchuan/ ~/.claude/skills/wangchuan/
 cp -r wangchuan/ ~/.openclaw/workspace/skills/wangchuan/
 # Codex
 cp -r wangchuan/ ~/.codex/skills/wangchuan/
-# Or let sync distribute automatically:
-wangchuan sync
+# Or let push distribute automatically:
+wangchuan push -y
 ```
 
 ## Setting up Git repo

package/skill/references/resource-crud.md

@@ -33,7 +33,7 @@ Skills and custom agents use the **same** flow — both are directories tracked
 - **Newly created** → NOT shared (skip to Step 2b).
 - **Existing** → check `~/.wangchuan/shared-registry.json` for the name with `kind:'skill'` or `kind:'agent'`.
 
-**Step 2a: Already-shared** — distribute + ask user to sync:
+**Step 2a: Already-shared** — distribute + ask user to push:
 ```bash
 # Example: shared skill "foo" modified in claude, also used by cursor and codebuddy
 cp -r ~/.claude/skills/foo/ ~/.cursor/skills/foo/
@@ -42,7 +42,7 @@ cp -r ~/.claude/skills/foo/ ~/.codebuddy/skills/foo/
 cp -r ~/.claude/agents/my-reviewer/ ~/.cursor/agents/my-reviewer/
 cp -r ~/.claude/agents/my-reviewer/ ~/.codebuddy/agents/my-reviewer/
 ```
-Tell user: "Changes saved locally and distributed to all agents. Run `wangchuan sync` to push to cloud." If user confirms: `wangchuan sync -y`
+Tell user: "Changes saved locally and distributed to all agents. Run `wangchuan push` to push to cloud." If user confirms: `wangchuan push -y`
 
 **Step 2b: New / agent-local** — resource stays in the current agent by default. Only distribute if user **explicitly asks** to share:
 ```bash
@@ -58,7 +58,7 @@ If user asks to share, present options:
 - **Specific agents** — copy only to selected
 - **No distribution** (default) — keep in current agent only
 
-After copying, tell user: "Changes saved locally. Run `wangchuan sync` to push to cloud." If user confirms: `wangchuan sync -y`
+After copying, tell user: "Changes saved locally. Run `wangchuan push` to push to cloud." If user confirms: `wangchuan push -y`
 
 ---
 
@@ -92,15 +92,15 @@ Options: **All agents** / **[individual agents]** / **Cancel**
 
 **Step 3: Execute.**
 
-**All agents**: `rm -rf` from every agent → unregister from shared-registry.json. Tell user: "Changes saved locally. Run `wangchuan sync` to push to cloud." If user confirms: `wangchuan sync -y`.
+**All agents**: `rm -rf` from every agent → unregister from shared-registry.json. Tell user: "Changes saved locally. Run `wangchuan push` to push to cloud." If user confirms: `wangchuan push -y`.
 ```bash
 rm -rf ~/.claude/skills/xxx/ ~/.cursor/skills/xxx/ ~/.codebuddy/skills/xxx/  # etc.
 jq '.entries |= map(select(.name != "xxx" or .kind != "skill"))' ~/.wangchuan/shared-registry.json > /tmp/wc-sr.json && mv /tmp/wc-sr.json ~/.wangchuan/shared-registry.json
-# Only sync when user confirms:
-wangchuan sync -y
+# Only push when user confirms:
+wangchuan push -y
 ```
 
-**Specific agents**: `rm -rf` from selected → unregister (demote). Tell user: "Changes saved locally. Run `wangchuan sync` to push to cloud." If user confirms: `wangchuan sync -y`. Remaining agents keep local copies.
+**Specific agents**: `rm -rf` from selected → unregister (demote). Tell user: "Changes saved locally. Run `wangchuan push` to push to cloud." If user confirms: `wangchuan push -y`. Remaining agents keep local copies.
 
 **Cancel**: no action.
 
@@ -117,7 +117,7 @@ MCP servers are JSON fields (`mcpServers`). Each agent's MCP config is **indepen
    - OpenClaw: `~/.openclaw/workspace/config/mcporter.json` → `mcpServers`
 2. Resource stays in the current agent by default. Only copy to other agents if the user **explicitly asks**.
    - If user asks to share: ask which agents → jq write to selected
-3. Tell user: "Changes saved locally. Run `wangchuan sync` to push to cloud." If user confirms: `wangchuan sync -y`
+3. Tell user: "Changes saved locally. Run `wangchuan push` to push to cloud." If user confirms: `wangchuan push -y`
 
 Example — manually copy a server to another agent:
 ```bash
@@ -129,7 +129,7 @@ jq --argjson srv "$SERVER_JSON" '.mcpServers["my-db"] = $srv' ~/.codebuddy/mcp.j
 1. List which agents have it: `jq -e '.mcpServers["xxx"]' <file>` per agent
 2. Ask user: All / Specific / Cancel
 3. Remove via `jq 'del(.mcpServers["xxx"])'` from selected agents
-4. Tell user: "Changes saved locally. Run `wangchuan sync` to push to cloud." If user confirms: `wangchuan sync -y`
+4. Tell user: "Changes saved locally. Run `wangchuan push` to push to cloud." If user confirms: `wangchuan push -y`
 
 **Note**: `jq del()` is the only way to remove an MCP server from an agent's config. There is no auto-propagation of deletions across agents.
 
@@ -145,12 +145,12 @@ Memory files are **per-agent** — no automatic cross-agent distribution.
    - **All**: `wangchuan memory broadcast <agent>`
    - **Specific**: `wangchuan memory copy <source> <target>`
    - **No**: keep local
-3. Tell user: "Changes saved locally. Run `wangchuan sync` to push to cloud." If user confirms: `wangchuan sync -y`
+3. Tell user: "Changes saved locally. Run `wangchuan push` to push to cloud." If user confirms: `wangchuan push -y`
 
 **Deleting:**
 1. `wangchuan memory list` to see who has what
 2. Ask user: All / Specific / Cancel
 3. `rm -f` selected agents' memory files
-4. Tell user: "Changes saved locally. Run `wangchuan sync` to push to cloud." If user confirms: `wangchuan sync -y`
+4. Tell user: "Changes saved locally. Run `wangchuan push` to push to cloud." If user confirms: `wangchuan push -y`
 
 **Note**: Memory filenames differ per agent — Claude uses `CLAUDE.md`, OpenClaw/CodeBuddy/WorkBuddy/Codex use `MEMORY.md`.

package/skill/references/sync-conflict.md

@@ -1,24 +1,24 @@
 # Pushing, Pulling, and Conflict Resolution
 
-## How sync works
+## How push and pull work
 
-`wangchuan sync` is **bidirectional**: pull first, then push. `wangchuan watch` is **pull-only** — it periodically pulls cloud changes but never pushes. Users must run `wangchuan sync` manually to push local changes.
+`wangchuan push` pushes local changes to cloud. `wangchuan pull` pulls cloud changes to local. Both are one-directional. Users run `pull` to get cloud data and `push` to upload local changes.
 
-**IMPORTANT: Pushing to cloud NEVER happens automatically.** The agent should only suggest `wangchuan sync -y` when the user explicitly asks to sync/push. After CRUD operations, inform the user that changes are saved locally and ask if they want to sync.
+**IMPORTANT: Pushing to cloud NEVER happens automatically.** The agent should only suggest `wangchuan push -y` when the user explicitly asks to push. After CRUD operations, inform the user that changes are saved locally and ask if they want to push.
 
-**IMPORTANT: Sync pushes each agent's files independently.** There is no cross-agent auto-distribution during sync. Each agent's resources (skills, MCP servers, custom agents) are synced to/from their own directory in the cloud repo. Sharing resources across agents is a separate, explicit user action.
+**IMPORTANT: Push sends each agent's files independently.** There is no cross-agent auto-distribution during push. Each agent's resources (skills, MCP servers, custom agents) are pushed to/from their own directory in the cloud repo. Sharing resources across agents is a separate, explicit user action.
 
-Flow of `sync`: **auto-snapshot** → fetch remote → if remote ahead → git pull → three-way merge → then stage + push local changes.
-Flow of `watch`: fetch remote → if remote ahead → git pull → restore to local → record unresolved conflicts.
+Flow of `push`: **auto-snapshot** → fetch remote → if remote ahead → git pull → three-way merge → then stage + push local changes.
+Flow of `pull`: fetch remote → if remote ahead → git pull → restore to local.
 
-**Note**: sync automatically creates a safety snapshot of the repo before every sync. This means you can always roll back to the pre-sync state via `wangchuan snapshot list` + `wangchuan snapshot restore`.
+**Note**: push automatically creates a safety snapshot of the repo before every push. This means you can always roll back to the pre-push state via `wangchuan snapshot list` + `wangchuan snapshot restore`.
 
-**Environment**: all sync/push/pull operations target the **current environment's branch only** (`cfg.environment`). After sync completes, always report the current environment name to the user (e.g. "Synced to cloud (environment: work)").
+**Environment**: all push/pull operations target the **current environment's branch only** (`cfg.environment`). After push completes, always report the current environment name to the user (e.g. "Pushed to cloud (environment: work)").
 
 ## Pushing memory to cloud
 
 1. **Preview** (optional): `wangchuan status -v`
-2. **Sync**: `wangchuan sync -y`
+2. **Push**: `wangchuan push -y`
 3. **Conflict resolution** (auto for `.md` files):
    - Non-overlapping edits → auto-merged ✅
    - Overlapping identical edits → deduplicated ✅
@@ -30,25 +30,29 @@ Flow of `watch`: fetch remote → if remote ahead → git pull → restore to lo
      (cloud changes)
      >>>>>>> REMOTE
      ```
-4. **If conflict markers written** → read file, show conflicts to user, ask how to resolve, edit to remove markers, `wangchuan sync -y` again.
+4. **If conflict markers written** → read file, show conflicts to user, ask how to resolve, edit to remove markers, `wangchuan push -y` again.
 5. **If no conflicts** → done.
 
 ```bash
 # Full push flow:
-wangchuan sync -y
+wangchuan push -y
 # Check for conflict markers:
 grep -l '<<<<<<< LOCAL' ~/.claude/CLAUDE.md ~/.openclaw/workspace/MEMORY.md ~/.codebuddy/MEMORY.md ~/.workbuddy/MEMORY.md ~/.codex/MEMORY.md 2>/dev/null
 ```
 
 ## Pulling memory from cloud
 
-Same command: `wangchuan sync -y`. Use `wangchuan sync -n` for dry-run preview first.
+```bash
+wangchuan pull
+```
+
+Use `wangchuan push -n` for a dry-run preview of what would be pushed.
 
 Conflict resolution is identical to pushing. Encrypted files are decrypted transparently during pull.
 
 **Pulling from a specific environment**: push/pull always targets the **current** environment's branch. To pull from a different env:
 ```bash
-wangchuan env switch <target-env>   # switches branch + auto-syncs (pulls target env's data)
+wangchuan env switch <target-env>   # switches branch + auto-pulls target env's data
 # When done, switch back:
 wangchuan env switch <original-env>
 ```
@@ -66,7 +70,7 @@ When user says "sync A's config to B", intent may be ambiguous.
 - **MCP**: read mcpServers from source, jq merge into target
 - **Custom agents**: `cp -r "${SRC_WS}/agents/"* "${DST_WS}/agents/"`
 
-After completing the local operations, tell user: "Changes saved locally. Run `wangchuan sync` to push to cloud." If user confirms: `wangchuan sync -y`
+After completing the local operations, tell user: "Changes saved locally. Run `wangchuan push` to push to cloud." If user confirms: `wangchuan push -y`
 
 Get workspace paths:
 ```bash
@@ -76,11 +80,11 @@ DST=$(jq -r '.profiles.default.<target>.workspacePath' ~/.wangchuan/config.json
 
 ## Files deleted from cloud
 
-When another machine deletes files from the cloud repo, the next pull (via `wangchuan sync` or watch daemon) automatically deletes them from the local workspace too. No confirmation needed — cloud is the single source of truth.
+When another machine deletes files from the cloud repo, the next pull (via `wangchuan pull`) automatically deletes them from the local workspace too. No confirmation needed — cloud is the single source of truth.
 
 All changes are preserved in git history. To recover accidentally deleted files:
 ```bash
 cd ~/.wangchuan/repo && git log --name-status -10   # find the deletion commit
 git checkout <hash>~1 -- <file-path>                 # restore the file
-wangchuan sync -y                                     # push it back
+wangchuan push -y                                     # push it back
 ```

package/skill/wangchuan-skill.sh

@@ -2,13 +2,11 @@
 # wangchuan-skill.sh — OpenClaw Skill script / OpenClaw Skill 脚本
 #
 # Usage (invoked by OpenClaw / Claude) / 用法（由 OpenClaw / Claude 调用）：
-#   wangchuan-skill.sh pull   [--agent openclaw|claude|gemini]
-#   wangchuan-skill.sh push   [--agent <name>] [--message "<msg>"]
-#   wangchuan-skill.sh status [--agent <name>]
-#   wangchuan-skill.sh diff   [--agent <name>]
-#   wangchuan-skill.sh list   [--agent <name>]
-#   wangchuan-skill.sh dump   [--agent <name>]
-#   wangchuan-skill.sh init   --repo <url>
+#   wangchuan-skill.sh pull     [--agent openclaw|claude|gemini]
+#   wangchuan-skill.sh push     [--agent <name>] [--dry-run] [--yes]
+#   wangchuan-skill.sh status   [--agent <name>] [--verbose]
+#   wangchuan-skill.sh init     --repo <url>
+#   wangchuan-skill.sh restore  --repo <url> --key <key>
 #
 # Environment variables / 环境变量：
 #   WANGCHUAN_DIR        Override wangchuan install path (default ~/wangchuan) / 可覆盖安装路径（默认 ~/wangchuan）
@@ -41,15 +39,15 @@ CMD="${1:-status}"
 shift || true   # Remove first arg, pass remaining / 移除第一个参数，剩余透传
 
 case "$CMD" in
-  pull|push|status|diff|list|dump|lang|sync|watch)
+  pull|push|status|diff|list|dump|lang)
     node "$BIN" "$CMD" "$@"
     ;;
-  init)
-    node "$BIN" init "$@"
+  init|restore)
+    node "$BIN" "$CMD" "$@"
     ;;
   *)
     echo "✖ Unknown command / 未知命令: $CMD" >&2
-    echo "  Available / 可用: pull | push | status | diff | list | dump | lang | sync | watch | init" >&2
+    echo "  Available / 可用: pull | push | status | diff | list | dump | lang | init" >&2
     exit 1
     ;;
 esac