cc-mirror 1.3.0 → 1.4.0

package/dist/skills/orchestration/SKILL.md

@@ -154,6 +154,7 @@ Before anything, sense the vibe:
 `Write` `Edit` `Glob` `Grep` `Bash` `WebFetch` `WebSearch` `LSP`
 
 **Coordination tools you USE DIRECTLY:**
+
 - `Read` — see guidelines below
 - `TaskCreate`, `TaskUpdate`, `TaskGet`, `TaskList` — task management
 - `AskUserQuestion` — clarify scope with the user
@@ -196,6 +197,7 @@ Before anything, sense the vibe:
 9. **Celebrate** → Mark the wins
 
 **The key distinction:**
+
 - Quick reads for coordination (1-2 files) → ✅ You do this
 - Comprehensive reading/analysis (3+ files) → ❌ Spawn an agent
 - Skill references → ✅ ALWAYS you (never delegate)
@@ -373,10 +375,10 @@ User gets: Complete understanding, not just a surface answer. Impressed.
 
 **Scale agents to the work:**
 
-| Complexity | Agents |
-|------------|--------|
-| Quick lookup, simple fix | 1-2 agents |
-| Multi-faceted question | 2-3 parallel agents |
+| Complexity                 | Agents                  |
+| -------------------------- | ----------------------- |
+| Quick lookup, simple fix   | 1-2 agents              |
+| Multi-faceted question     | 2-3 parallel agents     |
 | Full feature, complex task | Swarm of 4+ specialists |
 
 The goal is thoroughness, not a quota. Match the swarm to the challenge.
@@ -615,17 +617,17 @@ This is your brand. It tells users they're in capable hands.
 
 ## 🚫 Anti-Patterns (FORBIDDEN)
 
-| ❌ Forbidden                      | ✅ Do This                           |
-| --------------------------------- | ------------------------------------ |
-| Exploring codebase yourself       | Spawn Explore agent                  |
-| Writing/editing code yourself     | Spawn general-purpose agent          |
-| Running bash commands yourself    | Spawn agent                          |
-| "Let me quickly..."               | Spawn agent                          |
-| "This is simple, I'll..."         | Spawn agent                          |
-| One agent at a time               | Parallel swarm                       |
-| Text-based menus                  | AskUserQuestion tool                 |
-| Cold/robotic updates              | Warmth and personality               |
-| Jargon exposure                   | Natural language                     |
+| ❌ Forbidden                   | ✅ Do This                  |
+| ------------------------------ | --------------------------- |
+| Exploring codebase yourself    | Spawn Explore agent         |
+| Writing/editing code yourself  | Spawn general-purpose agent |
+| Running bash commands yourself | Spawn agent                 |
+| "Let me quickly..."            | Spawn agent                 |
+| "This is simple, I'll..."      | Spawn agent                 |
+| One agent at a time            | Parallel swarm              |
+| Text-based menus               | AskUserQuestion tool        |
+| Cold/robotic updates           | Warmth and personality      |
+| Jargon exposure                | Natural language            |
 
 **Note:** Reading skill references, domain guides, and agent outputs for synthesis is NOT forbidden — that's coordination work.
 

package/dist/skills/task-manager/SKILL.md

@@ -0,0 +1,86 @@
+---
+name: task-manager
+description: CLI tool for managing team tasks outside of Claude Code's built-in tools. Use this skill when tasks need to be archived, bulk cleaned up, or when you need a high-level view of task dependencies. Triggers: (1) User asks to clean up or archive completed tasks, (2) Too many resolved tasks cluttering the list, (3) Need to visualize task dependencies, (4) Want to manage tasks via command line.
+---
+
+# Task Manager CLI
+
+You have access to `cc-mirror tasks` - a CLI for managing team tasks.
+
+## Auto-Detection
+
+The CLI auto-detects your context:
+
+- **Variant**: From `CLAUDE_CONFIG_DIR` environment variable
+- **Team**: From current working directory (git root folder name)
+
+No need to specify `--variant` or `--team` in most cases.
+
+## Commands
+
+```bash
+# List tasks
+cc-mirror tasks                    # Open tasks (default)
+cc-mirror tasks --status all       # All tasks
+
+# View dependencies
+cc-mirror tasks graph              # Visual dependency tree
+
+# Archive resolved tasks (preserves history)
+cc-mirror tasks archive --resolved --dry-run   # Preview
+cc-mirror tasks archive --resolved             # Execute
+
+# Delete permanently (no history)
+cc-mirror tasks clean --resolved --dry-run     # Preview
+cc-mirror tasks clean --resolved               # Execute
+
+# Single task operations
+cc-mirror tasks show <id>          # View details
+cc-mirror tasks archive <id>       # Archive one task
+cc-mirror tasks delete <id>        # Delete one task
+```
+
+## When to Use
+
+Use `AskUserQuestion` to confirm with the user:
+
+```
+When tasks are cluttering:
+  → "Archive resolved tasks?" (preserves in archive folder)
+  → "Delete resolved tasks?" (permanent removal)
+  → "Show task graph first?" (see dependencies)
+
+When user wants to see other projects/teams:
+  → "Which team?" (then use --team flag)
+  → "Which variant?" (then use --variant flag)
+  → "All teams?" (then use --all flag)
+```
+
+## Viewing Other Teams/Variants
+
+By default, commands target your current team. To view others:
+
+```bash
+# Different team in same variant
+cc-mirror tasks --team other-project
+
+# Different variant
+cc-mirror tasks --variant zai --team my-project
+
+# All teams in current variant
+cc-mirror tasks --all
+
+# All variants, all teams
+cc-mirror tasks --all-variants --all
+```
+
+Ask the user which team/variant they want to view using `AskUserQuestion`.
+
+## Archive vs Delete
+
+| Action  | Command              | Effect                                    |
+| ------- | -------------------- | ----------------------------------------- |
+| Archive | `archive --resolved` | Moves to `archive/` folder with timestamp |
+| Delete  | `clean --resolved`   | Permanently removes files                 |
+
+**Prefer archive** - it preserves task history for future reference.

package/dist/tui.mjs

@@ -2325,6 +2325,63 @@ var removeOrchestratorSkill = (configDir) => {
     return { status: "failed", message };
   }
 };
+var TASK_MANAGER_SKILL_NAME = "task-manager";
+var findBundledTaskManagerSkillDir = () => {
+  const thisFile = fileURLToPath(import.meta.url);
+  const thisDir = path6.dirname(thisFile);
+  const devPath = path6.join(thisDir, "..", "skills", TASK_MANAGER_SKILL_NAME);
+  if (fs5.existsSync(devPath)) return devPath;
+  const distPath = path6.join(thisDir, "skills", TASK_MANAGER_SKILL_NAME);
+  if (fs5.existsSync(distPath)) return distPath;
+  const distPath2 = path6.join(thisDir, "..", "skills", TASK_MANAGER_SKILL_NAME);
+  if (fs5.existsSync(distPath2)) return distPath2;
+  return null;
+};
+var installTaskManagerSkill = (configDir) => {
+  const sourceDir = findBundledTaskManagerSkillDir();
+  if (!sourceDir) {
+    return { status: "failed", message: "bundled task-manager skill not found" };
+  }
+  const skillsDir = path6.join(configDir, "skills");
+  const targetDir = path6.join(skillsDir, TASK_MANAGER_SKILL_NAME);
+  const markerPath = path6.join(targetDir, MANAGED_MARKER);
+  try {
+    ensureDir2(skillsDir);
+    if (fs5.existsSync(targetDir) && !fs5.existsSync(markerPath)) {
+      return { status: "skipped", message: "existing skill is user-managed", path: targetDir };
+    }
+    if (fs5.existsSync(targetDir)) {
+      fs5.rmSync(targetDir, { recursive: true, force: true });
+    }
+    copyDir(sourceDir, targetDir);
+    fs5.writeFileSync(
+      markerPath,
+      JSON.stringify({ managedBy: "cc-mirror", updatedAt: (/* @__PURE__ */ new Date()).toISOString() }, null, 2)
+    );
+    return { status: "installed", path: targetDir };
+  } catch (error) {
+    const message = error instanceof Error ? error.message : String(error);
+    return { status: "failed", message };
+  }
+};
+var removeTaskManagerSkill = (configDir) => {
+  const skillsDir = path6.join(configDir, "skills");
+  const targetDir = path6.join(skillsDir, TASK_MANAGER_SKILL_NAME);
+  const markerPath = path6.join(targetDir, MANAGED_MARKER);
+  if (!fs5.existsSync(targetDir)) {
+    return { status: "skipped", message: "skill not installed" };
+  }
+  if (!fs5.existsSync(markerPath)) {
+    return { status: "skipped", message: "skill is user-managed, not removing" };
+  }
+  try {
+    fs5.rmSync(targetDir, { recursive: true, force: true });
+    return { status: "removed", path: targetDir };
+  } catch (error) {
+    const message = error instanceof Error ? error.message : String(error);
+    return { status: "failed", message };
+  }
+};
 var spawnAsync = (cmd, args) => {
   return new Promise((resolve) => {
     const child = spawn3(cmd, args, { stdio: "pipe" });
@@ -2535,6 +2592,12 @@ var TeamModeStep = class {
     } else if (skillResult.status === "failed") {
       state.notes.push(`Warning: orchestrator skill install failed: ${skillResult.message}`);
     }
+    const taskSkillResult = installTaskManagerSkill(paths.configDir);
+    if (taskSkillResult.status === "installed") {
+      state.notes.push("Task manager skill installed");
+    } else if (taskSkillResult.status === "failed") {
+      state.notes.push(`Warning: task-manager skill install failed: ${taskSkillResult.message}`);
+    }
     const systemPromptsDir = path8.join(paths.tweakDir, "system-prompts");
     const copiedFiles = copyTeamPackPrompts(systemPromptsDir);
     if (copiedFiles.length > 0) {
@@ -3980,6 +4043,12 @@ var TeamModeUpdateStep = class {
     } else if (skillResult.status === "failed") {
       state.notes.push(`Warning: orchestrator skill removal failed: ${skillResult.message}`);
     }
+    const taskSkillResult = removeTaskManagerSkill(meta.configDir);
+    if (taskSkillResult.status === "removed") {
+      state.notes.push("Task manager skill removed");
+    } else if (taskSkillResult.status === "failed") {
+      state.notes.push(`Warning: task-manager skill removal failed: ${taskSkillResult.message}`);
+    }
   }
   patchCli(ctx) {
     const { state, meta, paths } = ctx;
@@ -4038,6 +4107,12 @@ var TeamModeUpdateStep = class {
     } else if (skillResult.status === "failed") {
       state.notes.push(`Warning: orchestrator skill install failed: ${skillResult.message}`);
     }
+    const taskSkillResult = installTaskManagerSkill(meta.configDir);
+    if (taskSkillResult.status === "installed") {
+      state.notes.push("Task manager skill installed");
+    } else if (taskSkillResult.status === "failed") {
+      state.notes.push(`Warning: task-manager skill install failed: ${taskSkillResult.message}`);
+    }
     const systemPromptsDir = path17.join(meta.tweakDir, "system-prompts");
     const copiedFiles = copyTeamPackPrompts(systemPromptsDir);
     if (copiedFiles.length > 0) {

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "cc-mirror",
-  "version": "1.3.0",
+  "version": "1.4.0",
   "type": "module",
   "description": "Create multiple isolated Claude Code variants with custom providers (Z.ai, MiniMax, OpenRouter, Claude Code Router)",
   "author": "Numman Ali",