kiku-agent 1.0.0 → 1.0.4

package/README.md

@@ -1,122 +1,293 @@
-# 🌸 Ju Jingyi (鞠婧祎) Portal & Kiku AI Agent
-
-> Full-stack fan archive, REST API Suite v1.0, and Autonomous AI Agent CLI powered by **LM Studio** (`gemma-4-e4b` / Gemma models).
-
----
-
-## 🚀 Kiku AI Agent CLI
-
-The **Kiku AI Agent** is an autonomous ReAct agent and specialized encyclopedia for Ju Jingyi. It connects to your local **LM Studio** instance and has access to catalog, web, media, filesystem, and music tools.
-
-### Global installation
-
-```bash
-npm install -g kiku-agent
-kiku-agent
-```
-
-The same package also installs the stdio MCP server command:
-
-```bash
-kiku-mcp
-```
-
-For a local checkout before the package is published, use `npm pack` followed by `npm install -g ./kiku-agent-1.0.0.tgz`.
-
-### ✨ Features
-- **LM Studio Integration**: Connects seamlessly to LM Studio's OpenAI-compatible server at `http://127.0.0.1:1234/v1` (with support for `gemma-4-e4b`, `gemma-3-4b`, and auto-detection of loaded models).
-- **Validated Tool Calling**: Schema validation, argument coercion, timeouts, duplicate-call caching, normalized results, and complete catalog/media integrations.
-- **Multilingual Support**: Automatically detects and responds in Thai, Chinese, or English based on user input language.
-- **Multi-turn Memory**: Retains conversation context with clear memory controls (`/clear`).
-- **Resilient ReAct Loop**: Supports OpenAI function calling as well as structured JSON tool parsing optimized for Gemma models.
-- **Graceful Diagnostics**: Displays real-time connection status, loaded model names, and fallback search modes if LM Studio is offline.
-- **Clean Thinking Display**: Shows reasoning process without markdown artifacts for better readability.
-
----
-
-## 🛠️ LM Studio Quick Setup
-
-1. Open **LM Studio** on your computer.
-2. In the Search tab, download or load **Gemma** (e.g. `gemma-4-e4b`, `gemma-3-4b-it`, `gemma-2-9b-it`).
-3. Click the **Developer / Local Server** tab (icon with `<->`) on the left bar.
-4. Select your loaded Gemma model and click **Start Server** on port `1234` (`http://localhost:1234`).
-5. In this terminal, run:
-   ```bash
-   npm run kiku-agent
-   ```
-
----
-
-## 💻 CLI Usage & Commands
-
-### Interactive Chat Mode
-```bash
-npm run kiku-agent
-```
-
-### One-shot Question Query
-```bash
-npm run kiku-agent -- --query "ผลงานซีรีส์ปี 2026 มีเรื่องอะไรบ้างและรับบทเป็นใคร"
-```
-
-### Specify Custom Model or LM Studio URL
-```bash
-npm run kiku-agent -- --model gemma-4-e4b --url http://127.0.0.1:1234/v1
-```
-
-### View Catalog Overview Statistics
-```bash
-npm run kiku-agent -- --stats
-```
-
-### View All Registered Tools
-```bash
-npm run kiku-agent -- --tools
-```
-
-### Interactive Chat Slash Commands
-- `/help` — Display list of commands
-- `/tools` — View all tool names and parameter descriptions
-- `/details [Qn]` — Expand the recorded thinking and tool results for a question (`F2` opens the latest)
-- `/stats` — Print summary counts of dramas, songs, brands, and awards
-- `/model` — Inspect active models currently loaded in LM Studio
-- `/thinking [show|hide]` — Toggle reasoning process visibility
-- `/save` — Export session transcript to Markdown file
-- `/clear` — Reset conversation history
-- `/exit` — Exit the agent
-
----
-
-## 📚 Available API Tools in Kiku Agent
-
-| # | Tool Name | Description |
-|---|---|---|
-| 1 | `get_profile_and_fandom` | Bio, Thai/Chinese/Pinyin names, Miju (蜜橘) fan community, slogans, socials |
-| 2 | `get_quotes` | Inspiring quotes with Thai translation and original Chinese text |
-| 3 | `search_works` | Filter filmography by year (2013-2026), category, keywords |
-| 4 | `get_work_detail` | Detailed synopsis, roles, ratings, streaming platform, and OST |
-| 5 | `get_character_roles` | Portrayed roles (Yang Caiwei, Lu Wuyi, Bai Suzhen, Han Yunxi, etc.) |
-| 6 | `search_discography` | Deduplicated music catalog, singles, OSTs, and albums |
-| 7 | `get_music_videos` | MVs, stage performances, dance practices, and special stages |
-| 8 | `get_brand_endorsements`| Commercial endorsements, luxury brands (PRSR, FRED), Milan fashion week |
-| 9 | `get_career_timeline` | Milestone eras from SNH48 debut (2013) to solo studio to present |
-| 10 | `get_awards` | Honors in music, television acting, and fashion influence |
-| 11 | `get_social_feed` | Recent Weibo & Douyin social media updates and metrics |
-| 12 | `get_gallery_photos` | High-res fashion sets, red carpet, photoshoot, and drama stills |
-| 13 | `global_search` | Universal fuzzy search across all entities in the database |
-| 14 | `web_search` | Live internet search for up-to-date news and info |
-| 15 | `fetch_webpage` | Extract and read content from live URLs |
-| 16 | `fetch_api` | Query the official Ju Jingyi REST API for structured JSON data |
-| 17 | `get_system_stats` | High-level summary metrics of the entire Ju Jingyi archive |
-
----
-
-## 🌐 Web App & API Development
-
-```bash
-# Start frontend dev server
-npm run dev
-
-# Run API verification tests
-powershell -ExecutionPolicy Bypass -File test_api.ps1
-```
+# 🌸 Ju Jingyi (鞠婧祎) Portal & Kiku AI Agent
+
+> **New CLI capabilities:** `/subagent <task>` runs a bounded Ju Jingyi research worker, while `/skill learn <instruction>` teaches Kiku persistent response preferences.
+
+> Full-stack fan archive, REST API Suite v1.0, autonomous AI Agent CLI, and MCP server powered by **LM Studio** (`google/gemma-4-e4b` / Gemma models).
+
+---
+
+## 🚀 Kiku AI Agent CLI
+
+The **Kiku AI Agent** is an autonomous ReAct agent and specialized encyclopedia for Ju Jingyi. It connects to your local **LM Studio** instance and has access to catalog, web, media, filesystem, and music tools.
+
+### Global installation
+
+```bash
+npm install -g kiku-agent
+kiku-agent
+```
+
+The global package intentionally excludes native audio build dependencies, so Windows users do not need Python, Visual Studio Build Tools, `node-gyp`, or FFmpeg to install the core CLI. On a new machine, run `kiku-agent` and then `/doctor`; if LM Studio or a chat model is unavailable, Kiku starts in catalog-only mode instead of crashing.
+
+If Windows is upgrading from an older failed install and reports `EBUSY` for `ffmpeg-static`, close every running Kiku/Node terminal and use a fresh PowerShell window:
+
+```powershell
+npm uninstall -g kiku-agent
+$stale = Join-Path (npm root -g) 'kiku-agent'
+if (Test-Path $stale) { Remove-Item -Recurse -Force $stale }
+npm cache verify
+npm install -g kiku-agent
+```
+
+Node 20 remains supported by the published core CLI. Ink 7 is development-only and is not installed globally; the Node 22 engine warnings from `ink`, `cli-truncate`, and `slice-ansi` should therefore no longer appear.
+
+Kiku never needs Administrator privileges. Sessions are stored in `%APPDATA%\\kiku-agent` on Windows, not in the current directory or under `Program Files`. If an older release still tries to create `.kiku_sessions` in a protected folder, install version 1.0.2 or newer and remove any old `KIKU_DATA_DIR` override:
+
+```powershell
+[Environment]::SetEnvironmentVariable('KIKU_DATA_DIR', $null, 'User')
+$env:KIKU_DATA_DIR = $null
+npm uninstall -g kiku-agent
+npm cache verify
+npm install -g kiku-agent@latest
+kiku-agent
+```
+
+The same package also installs the stdio MCP server command:
+
+```bash
+kiku-mcp
+```
+
+For a local checkout before the package is published, use `npm pack` followed by `npm install -g ./kiku-agent-1.0.4.tgz`.
+
+### ✨ Features
+- **LM Studio Integration**: Connects seamlessly to LM Studio's OpenAI-compatible server at `http://127.0.0.1:1234/v1` (default model `google/gemma-4-e4b`, with auto-detection of loaded models).
+- **Validated Tool Calling**: Schema validation, argument coercion, timeouts, duplicate-call caching, normalized results, and 38 catalog/media/web/filesystem/music tools.
+- **Music Player**: Search YouTube, stream audio locally through ffmpeg, and control a queue with play/pause/skip commands.
+- **Multilingual Support**: Automatically detects and responds in Thai, Chinese, or English based on user input language.
+- **Multi-turn Memory**: Retains conversation context with memory inspection and clear controls.
+- **Resilient ReAct Loop**: Supports OpenAI function calling as well as structured JSON tool parsing optimized for Gemma models.
+- **Graceful Diagnostics**: Displays real-time connection status, loaded model names, and fallback search modes if LM Studio is offline.
+- **Interactive TUI**: Optional Ink-based terminal UI (`npm run kiku-ink`) with themes, thinking panels, and tool result rendering.
+
+---
+
+## 🛠️ LM Studio Quick Setup
+
+1. Open **LM Studio** on your computer.
+2. In the Search tab, download or load **`google/gemma-4-e4b`** (quantization: `Q4_K_M`, `Q6_K`, or `Q8_0`).
+3. Click the **Developer / Local Server** tab (icon with `<->`) on the left bar.
+4. Select your loaded Gemma model and click **Start Server** on port `1234` (`http://localhost:1234`).
+5. In this terminal, run:
+   ```bash
+   npm run kiku-agent
+   ```
+
+---
+
+## 💻 CLI Usage
+
+### Interactive Chat Mode
+```bash
+npm run kiku-agent
+```
+
+### Launch with Specific Cloud Provider & Model
+```bash
+npm run kiku-agent -- --provider openai --model gpt-4o-mini
+npm run kiku-agent -- --provider gemini --model gemini-2.5-flash
+npm run kiku-agent -- --provider groq --model llama-3.1-8b-instant
+npm run kiku-agent -- --provider openrouter --model google/gemini-2.5-flash
+```
+
+### One-shot Question Query
+```bash
+npm run kiku-agent -- --query "ผลงานซีรีส์ปี 2026 มีเรื่องอะไรบ้างและรับบทเป็นใคร"
+```
+
+### Specify Custom Model or Base URL
+```bash
+npm run kiku-agent -- --model google/gemma-4-e4b-Q4_K_M --url http://127.0.0.1:1234/v1
+```
+
+### CLI Flags
+| Flag | Description |
+|---|---|
+| `-h`, `--help` | Show usage help |
+| `-P`, `--provider <name>` | Select active provider (`lmstudio`, `openai`, `gemini`, `groq`, `deepseek`, `openrouter`, `custom`) |
+| `-k`, `--api-key <key>` | Specify API key for cloud provider |
+| `-m`, `--model <id>` | Override model ID (e.g. `gpt-4o-mini`, `google/gemma-4-e4b-Q4_K_M`) |
+| `-u`, `--url <url>` | Override provider base URL |
+| `-t`, `--tools` | List all registered tools |
+| `-s`, `--stats` | Print catalog overview statistics |
+| `-v`, `--verbose` | Verbose mode with reasoning shown |
+| `--thinking [show\|hide]` | Toggle reasoning visibility |
+| `-q`, `--query "<text>"` | One-shot query mode |
+
+### Interactive Chat Slash Commands
+
+Run `/help` inside a session for the full list. Commonly used:
+
+- `/provider` — View active AI provider status, model, URL, and API key status
+- `/provider list` — View all 7 supported AI providers and readiness status
+- `/provider use <name> [model]` — Switch provider and model in a single command
+- `/provider models [name]` — List curated active cost-effective models with pricing & context window
+- `/provider info <name>` — Display developer documentation links, API references, and required ENV keys
+- `/provider test [name]` — Test live endpoint connection and measure latency in milliseconds
+- `/provider setkey <key>` — Dynamically set or update API key at runtime
+- `/model` — Inspect active models currently loaded in LM Studio or cloud provider
+- `/model list` — List loaded models from server alongside recommended models
+- `/model recommend` — Show recommended active models for current provider
+- `/model use <model-id>` — Switch runtime model without restarting the CLI
+- `/model reconnect` — Recheck server and leave catalog-only mode after a chat model is loaded
+- `/doctor` — Check Node, writable app data, LM Studio, model availability, FFmpeg, speaker, and yt-dlp
+- `/tools` — View all tool names and parameter descriptions
+- `/details [Qn]` — Expand recorded thinking and tool results for a question
+- `/stats` — Print summary counts of dramas, songs, brands, and awards
+- `/thinking [show|hide]` — Toggle reasoning process visibility
+- `/save` — Export session transcript to Markdown file
+- `/history` — Review conversation history
+- `/theme` — Switch terminal color theme
+- `/subagent <task>` — Send a focused Ju Jingyi research subtask to a bounded worker with catalog and web-research tools
+- `/research <question>` — Coordinate catalog, current-web, and verification workers in parallel, then synthesize one sourced answer
+- `/verify <claim>` — Classify a claim as verified, unverified, conflicting, or false
+- `/sources` — Show the source URLs captured by the latest research run
+- `/skill learn <instruction>` — Teach Kiku a persistent response preference, such as “ตอบภาษาไทยแบบสั้น กระชับ”
+- `/skill list`, `/skill del <id>`, `/skill on|off`, `/skill clear` — Review and control learned response skills
+- Music controls: `/play <song>`, `/stop`, `/pause`, `/resume`, `/next`, `/prev`, `/queue`, `/add <song>`, `/remove <n>`, `/clearqueue`, `/mute`, `/unmute`, `/current`
+- `/clear` — Reset conversation history
+- `/exit` — Exit the agent
+
+---
+
+## 📚 Available API Tools in Kiku Agent (38)
+
+**Catalog**
+| Tool Name | Description |
+|---|---|
+| `get_profile_and_fandom` | Bio, Thai/Chinese/Pinyin names, Miju (蜜橘) fan community, slogans, socials |
+| `get_quotes` | Inspiring quotes with Thai translation and original Chinese text |
+| `search_works` | Filter filmography by year (2013–2026), category, keywords |
+| `get_work_detail` | Detailed synopsis, roles, ratings, streaming platform, and OST |
+| `get_character_roles` | Portrayed roles (Yang Caiwei, Lu Wuyi, Bai Suzhen, Han Yunxi, etc.) |
+| `search_discography` | Deduplicated music catalog, singles, OSTs, and albums |
+| `get_music_videos` | MVs, stage performances, dance practices, and special stages |
+| `get_media_links` | Locate official media links by song/drama/stage/MV name |
+| `get_brand_endorsements` | Commercial endorsements, luxury brands (PRSR, FRED), Milan fashion week |
+| `get_career_timeline` | Milestone eras from SNH48 debut (2013) to solo studio to present |
+| `get_awards` | Honors in music, television acting, and fashion influence |
+| `get_social_feed` | Recent Weibo & Douyin social media updates and metrics |
+| `get_gallery_photos` | High-res fashion sets, red carpet, photoshoot, and drama stills |
+| `global_search` | Universal fuzzy search across all entities in the database |
+
+**Web & Network**
+| Tool Name | Description |
+|---|---|
+| `web_search` | Live internet search for up-to-date news and info |
+| `web_extract` | Extract readable article content from live pages |
+| `fetch_webpage` | Fetch raw page content from live URLs |
+| `fetch_json` | Fetch and parse JSON from any endpoint |
+| `screenshot_webpage` | Capture a webpage screenshot |
+| `fetch_api` | Query the official Ju Jingyi REST API for structured JSON data |
+
+**Music Player (local audio via ffmpeg)**
+| Tool Name | Description |
+|---|---|
+| `search_youtube` | Search YouTube for songs and videos |
+| `play_music` | Play a song by name via yt-dlp/ytdl-core + ffmpeg |
+| `pause_music` / `resume_music` / `stop_music` | Playback control |
+| `next_track` / `prev_track` | Skip within the queue |
+| `add_to_queue` / `remove_from_queue` / `get_queue` / `clear_queue` | Queue management |
+| `get_current_track` / `get_music_status` | Current playback state |
+
+**Filesystem & System**
+| Tool Name | Description |
+|---|---|
+| `list_directory` | List files in an allowed local directory |
+| `read_local_file` | Read a local text file safely |
+| `get_system_stats` | High-level summary metrics of the entire Ju Jingyi archive |
+
+---
+
+## 🤖 Multi-Provider Cloud & Local Models
+
+`kiku-agent` natively supports local LM Studio inference alongside top Cloud AI Providers with official endpoints, active cost-effective models, and tool-calling capabilities:
+
+| Provider | Recommended Active Models | Pricing / Cost Tier | Doc & API Reference | Required ENV Variable |
+|---|---|---|---|---|
+| **LM Studio** (Local) | `google/gemma-4-e4b` *(Q4_K_M, Q6_K, Q8_0)* | **100% Free / Local** (Zero token cost) | [Docs](https://lmstudio.ai/docs) • [API Ref](https://lmstudio.ai/docs/api/local-server) | `LMSTUDIO_BASE_URL` (default: `http://127.0.0.1:1234/v1`) |
+| **OpenAI** | `gpt-4o-mini` *(default)*, `gpt-4o`, `o3-mini` | **Ultra-Cheap** (~$0.15 / 1M input) | [Docs](https://platform.openai.com/docs) • [API Ref](https://platform.openai.com/docs/api-reference) | `OPENAI_API_KEY` |
+| **Google Gemini** | `gemini-2.5-flash` *(default)*, `gemini-2.5-flash-lite`, `gemini-2.5-pro`, `gemini-3.1-pro` | **Ultra-Cheap** ($0.10 / 1M in, 1M-2M context) | [Docs](https://ai.google.dev/docs) • [API Ref](https://ai.google.dev/api) | `GEMINI_API_KEY` |
+| **Groq** | `llama-3.3-70b-versatile` *(default)*, `llama-3.1-8b-instant`, `gpt-oss-120b` | **LPU Ultra-Fast** ($0.05 - $0.59 / 1M in) | [Docs](https://console.groq.com/docs/quickstart) • [API Ref](https://console.groq.com/docs/api-reference) | `GROQ_API_KEY` |
+| **DeepSeek** | `deepseek-chat` *(default, V3)*, `deepseek-reasoner` *(R1 CoT)* | **Ultra-Low Cost** ($0.14 - $0.27 / 1M in) | [Docs](https://api-docs.deepseek.com) • [API Ref](https://api-docs.deepseek.com/api/create-chat-completion) | `DEEPSEEK_API_KEY` |
+| **OpenRouter** | `google/gemini-2.5-flash`, `deepseek/deepseek-chat`, `deepseek/deepseek-r1`, `meta-llama/llama-3.3-70b-instruct`, `qwen/qwen-2.5-72b-instruct`, `:free` | **Flexible Aggregator** (Free & Paid) | [Docs](https://openrouter.ai/docs/quick-start) • [API Ref](https://openrouter.ai/docs/api_reference/overview) | `OPENROUTER_API_KEY` |
+| **Custom Gateway** | Any OpenAI-compatible server (vLLM, Ollama, TGI, Azure) | **Custom** | [OpenAI Spec](https://platform.openai.com/docs/api-reference) | `LLM_BASE_URL`, `LLM_API_KEY` |
+
+```bash
+# Launch with a specific cloud provider and model
+npm run kiku-agent -- --provider openai --model gpt-4o-mini
+npm run kiku-agent -- --provider gemini --model gemini-2.5-flash
+npm run kiku-agent -- --provider groq --model llama-3.1-8b-instant
+npm run kiku-agent -- --provider openrouter --model google/gemini-2.5-flash
+
+# Manage providers interactively within the agent CLI
+/provider list                                     # View all 7 providers and key status
+/provider use groq llama-3.1-8b-instant            # Switch provider and model in one command
+/provider models gemini                            # List curated cost-effective models with pricing & context
+/provider info openai                              # Show developer documentation & API details
+/provider test groq                                # Test live endpoint connection and measure latency
+/provider setkey groq gsk-...                      # Set or update API key dynamically at runtime
+/model recommend                                   # View recommended models for active provider
+/model use gpt-4o-mini                             # Switch active model on current provider
+```
+
+---
+
+## 🔌 Model Context Protocol (MCP)
+
+### 1. MCP Server (`kiku-mcp`)
+The package ships a Model Context Protocol server exposing the catalog as tools, resources, and prompts:
+```bash
+# stdio transport (Claude Desktop, Cursor, ...)
+npm run kiku-mcp
+```
+The production deployment exposes a stateless Streamable HTTP endpoint at `https://jujingyi-kikuuu.vercel.app/mcp` (`POST /mcp`).
+
+### 2. MCP Client in `kiku-agent`
+`kiku-agent` can connect to external MCP servers (Streamable HTTP, Stdio) and dynamically load their tools:
+```bash
+# Connect to an MCP server from CLI:
+/mcp connect kiku-remote https://jujingyi-kikuuu.vercel.app/mcp
+/mcp connect filesystem stdio npx -y @modelcontextprotocol/server-filesystem ./
+/mcp tools
+```
+
+---
+
+## ⚙️ Environment Variables
+
+Copy `.env.example` → `.env` (parsed automatically when the CLI runs; only unset variables are filled):
+
+| Variable | Purpose |
+|---|---|
+| `LLM_PROVIDER` | Active provider override (`openai`, `openrouter`, `groq`, `deepseek`, `gemini`, `lmstudio`, `custom`) |
+| `OPENAI_API_KEY` | OpenAI API Key |
+| `OPENROUTER_API_KEY` | OpenRouter API Key |
+| `GROQ_API_KEY` | Groq Cloud API Key |
+| `DEEPSEEK_API_KEY` | DeepSeek API Key |
+| `GEMINI_API_KEY` | Google Gemini API Key |
+| `LMSTUDIO_BASE_URL` | LM Studio endpoint (default `http://127.0.0.1:1234/v1`) |
+| `KIKU_YT_COOKIES` | Path to a `cookies.txt` file for YouTube playback |
+| `KIKU_YT_BROWSER` | Import cookies from a browser (`chrome`, `edge`, `firefox`) |
+| `KIKU_YTDLP` | Path/command for yt-dlp binary (default `yt-dlp`) |
+| `KIKU_YTDLP_CLIENT` | YouTube player client override (`tv`, `ios`, `web`) |
+
+> `.env` is gitignored — never commit it. The checked-in `.env.example` contains placeholders only.
+
+---
+
+## 🌐 Web App & API Development
+
+```bash
+# Start frontend dev server (API v1 works in-process)
+npm run dev
+
+# Production build
+npm run build
+
+# Run unit tests (agent + MCP, no LM Studio required)
+npm test
+
+# Run API verification tests against localhost:3000
+powershell -ExecutionPolicy Bypass -File test_api.ps1 -BaseHost localhost:3000
+```

package/package.json

@@ -1,76 +1,67 @@
 {
-  "name": "kiku-agent",
-  "version": "1.0.0",
-  "description": "Kiku AI terminal agent and MCP server for Ju Jingyi knowledge, media, and career data.",
-  "type": "module",
-  "bin": {
-    "kiku-agent": "server/agent/cli.js",
-    "kiku-mcp": "server/mcp/stdio.js"
-  },
-  "exports": {
-    ".": "./server/agent/agentCore.js",
-    "./mcp": "./server/mcp/server.js",
-    "./tools": "./server/agent/tools.js"
-  },
-  "files": [
-    "server/agent/*.js",
-    "server/mcp/*.js",
-    "server/mcp/data/*.js",
-    "server/mcp/**/*.json",
-    "src/data/**/*.js",
-    "README.md"
-  ],
-  "engines": {
-    "node": ">=20.0.0"
-  },
-  "keywords": [
-    "kiku",
-    "ju-jingyi",
-    "cli",
-    "ai-agent",
-    "mcp",
-    "model-context-protocol",
-    "lm-studio"
-  ],
-  "repository": {
-    "type": "git",
-    "url": "git+https://github.com/ThemeHackers/jujingyi_kikuuu.git"
-  },
-  "homepage": "https://jujingyi-kikuuu.vercel.app/",
-  "bugs": {
-    "url": "https://github.com/ThemeHackers/jujingyi_kikuuu/issues"
-  },
-  "license": "UNLICENSED",
-  "publishConfig": {
-    "access": "public"
-  },
-  "scripts": {
+  "name": "kiku-agent",
+  "version": "1.0.4",
+  "description": "Kiku AI terminal agent and MCP server for Ju Jingyi knowledge, media, and career data.",
+  "type": "module",
+  "bin": {
+    "kiku-agent": "server/agent/cli.js",
+    "kiku-mcp": "server/mcp/stdio.js"
+  },
+  "exports": {
+    ".": "./server/agent/agentCore.js",
+    "./mcp": "./server/mcp/server.js",
+    "./tools": "./server/agent/tools.js"
+  },
+  "files": [
+    "server/agent/*.js",
+    "server/mcp/*.js",
+    "server/mcp/data/*.js",
+    "server/mcp/**/*.json",
+    "src/data/**/*.js",
+    "README.md"
+  ],
+  "engines": {
+    "node": ">=20.0.0"
+  },
+  "keywords": [
+    "kiku",
+    "ju-jingyi",
+    "cli",
+    "ai-agent",
+    "mcp",
+    "model-context-protocol",
+    "lm-studio"
+  ],
+  "repository": {
+    "type": "git",
+    "url": "git+https://github.com/ThemeHackers/jujingyi_kikuuu.git"
+  },
+  "homepage": "https://jujingyi-kikuuu.vercel.app/",
+  "bugs": {
+    "url": "https://github.com/ThemeHackers/jujingyi_kikuuu/issues"
+  },
+  "license": "UNLICENSED",
+  "publishConfig": {
+    "access": "public"
+  },
+  "scripts": {
     "dev": "vite",
     "build": "vite build",
-    "preview": "vite preview",
-    "kiku-agent": "node server/agent/cli.js",
-    "kiku-mcp": "node server/mcp/stdio.js",
-    "kiku-ink": "npx tsx server/agent/ink/main.jsx",
-    "test": "node --test server/agent/tests/*.test.js server/mcp/tests/*.test.js",
-    "prepublishOnly": "npm test"
+    "preview": "vite preview",
+    "kiku-agent": "node server/agent/cli.js",
+    "kiku-mcp": "node server/mcp/stdio.js",
+    "kiku-ink": "npx tsx server/agent/ink/main.jsx",
+    "test": "node --test server/agent/tests/*.test.js server/mcp/tests/*.test.js",
+    "prepublishOnly": "npm test"
   },
   "dependencies": {
-    "@distube/ytdl-core": "^4.16.12",
     "@modelcontextprotocol/sdk": "^1.30.0",
     "clsx": "^2.1.1",
-    "ffmpeg-static": "^5.3.0",
-    "fluent-ffmpeg": "^2.1.3",
     "framer-motion": "^12.4.7",
-    "ink": "^7.1.1",
-    "ink-select-input": "^6.2.0",
-    "ink-spinner": "^5.0.0",
-    "ink-text-input": "^6.0.0",
     "lucide-react": "^1.16.0",
     "react": "^19.0.0",
     "react-dom": "^19.0.0",
-    "speaker": "^0.5.5",
     "tailwind-merge": "^3.0.2",
-    "yt-search": "^2.13.1",
     "zod": "^4.4.3"
   },
   "devDependencies": {
@@ -78,8 +69,16 @@
     "@types/react": "^19.0.10",
     "@types/react-dom": "^19.0.4",
     "@vitejs/plugin-react": "^4.3.4",
+    "ink": "^7.1.1",
+    "ink-select-input": "^6.2.0",
+    "ink-spinner": "^5.0.0",
+    "ink-text-input": "^6.0.0",
     "tailwindcss": "^4.0.9",
     "tsx": "^4.23.12",
     "vite": "^6.2.0"
+  },
+  "allowScripts": {
+    "esbuild@0.25.12": true,
+    "esbuild@0.28.2": true
   }
 }