mcp server
claude-plugins
Claude Code plugin marketplace for the n24q02m MCP servers -- install web search, memory, Notion, email, Telegram, Godot, code review, and image/video tools in one command.
Set up
Runs on your computer. Pick your agent and operating system.
This project does not publish an install method we can turn into a config. Follow the README instead.
README
synced Oct 17, 2026n24q02m Claude Plugins
| Project | Tagline | Tag |
|---|---|---|
| agent-chat-plugin | Peer AI agents chat in a shared folder — no human relay, no orchestrator, wor... | Tooling |
| better-code-review-graph | Knowledge graph for token-efficient code reviews -- semantic search and call-... | MCP |
| better-drive | 2-way Google Drive sync with .driveignore filter — rclone engine, Windows tray | Tooling |
| better-email-mcp | IMAP/SMTP email for AI agents -- read, send, organize folders, and manage att... | MCP |
| better-godot-mcp | Composite MCP server for Godot Engine -- 17 composite tools for AI-assisted g... | MCP |
| better-notion-mcp | Markdown-first Notion for AI agents -- pages, databases, blocks, and comments... | MCP |
| better-semantic-release | Drop-in python-semantic-release fork with built-in release-safety guards (orp... | Tooling |
| better-telegram-mcp | Telegram for AI agents -- messages, chats, media, and contacts across both bo... | MCP |
| better-workspace-mcp | Google Workspace MCP server (Docs/Drive/Calendar/Gmail/Sheets/Slides/Tasks/Ch... | MCP |
| claude-plugins | Claude Code plugin marketplace for the n24q02m MCP servers -- install web sea... | Marketplace |
| imagine-mcp | Image and video understanding + generation for AI agents -- across Gemini, Op... | MCP |
| jules-task-archiver | Chrome Extension for bulk operations on Jules tasks via batchexecute API -- a... | Tooling |
| mcp-core | Shared foundation for building MCP servers -- Streamable HTTP transport, OAut... | MCP |
| mnemo-mcp | Persistent AI memory with hybrid search and embedded sync. Open, free, unlimi... | MCP |
| qwen3-embed | Lightweight Qwen3 text embedding and reranking via ONNX Runtime and GGUF | Library |
| skret | Secrets without the server. | CLI |
| tacet | A self-distilling neuro-symbolic cascade that amortises LLM cost across knowl... | Tooling |
| web-core | Shared web infrastructure package for search, scraping, HTTP security, and st... | Library |
| wet-mcp | Open-source MCP server for AI agents: web search, content extraction, and lib... | MCP |
Table of contents
9 MCP servers and a portable Agent Chat coordination plugin for AI coding agents.
Full documentation: mcp.n24q02m.com — setup and tool references for all 9 servers, Agent Chat coordination, the mcp-core foundation, and Fastretrieval. Libraries and coordination plugins do not gain an MCP transport by appearing in the same documentation site.
Install
/plugin marketplace add n24q02m/claude-plugins
/plugin install <plugin-name>@n24q02m-plugins
Or browse all plugins: run /plugin and go to the Discover tab.
Plugins
| Plugin | Category | Description | Config (env / userConfig) |
|---|---|---|---|
| wet-mcp | Research | Web search, content extraction, library docs, media download | All optional: JINA_AI_API_KEY, GEMINI_API_KEY, GOOGLE_VERTEX_EXPRESS_API_KEY, OPENAI_API_KEY, COHERE_API_KEY, GITHUB_TOKEN |
| mnemo-mcp | Productivity | Persistent AI memory across sessions | All optional: JINA_AI_API_KEY, GEMINI_API_KEY, GOOGLE_VERTEX_EXPRESS_API_KEY, OPENAI_API_KEY, COHERE_API_KEY |
| better-notion-mcp | Productivity | Notion API — 11 composite tools, ~95% coverage | NOTION_TOKEN (required) |
| better-email-mcp | Communication | Email IMAP/SMTP — multi-account | EMAIL_CREDENTIALS (required) |
| better-telegram-mcp | Communication | Telegram dual-mode (Bot API + MTProto) — messages, chats, media | TELEGRAM_BOT_TOKEN (optional; required for bot mode) |
| better-godot-mcp | Development | Godot Engine — 17 composite tools for game dev | GODOT_PATH, GODOT_PROJECT_PATH (both optional) |
| better-code-review-graph | Development | Knowledge graph for token-efficient code reviews | All optional: JINA_AI_API_KEY, GEMINI_API_KEY, GOOGLE_VERTEX_EXPRESS_API_KEY, OPENAI_API_KEY, COHERE_API_KEY |
| imagine-mcp | Multimodal | Image/video understanding + generation across Gemini, OpenAI, Grok | All optional: XAI_API_KEY (default provider), GEMINI_API_KEY, GOOGLE_VERTEX_EXPRESS_API_KEY, OPENAI_API_KEY |
| better-workspace-mcp | Productivity | Google Workspace — 10 domains plus local time helpers, config, help, and multi-account OAuth | GOOGLE_OAUTH_CLIENT_ID, GOOGLE_OAUTH_CLIENT_SECRET (both required) |
| agent-chat-plugin | Development | Portable CLI/Skill peer coordination through Markdown/JSON messages, tasks, leases, and path locks; not an MCP server | Shared folder and agent identity; no provider API keys |
Fastretrieval is the retrieval library used by Wet, Mnemo, and CRG. Install it as a Python dependency, not as a marketplace plugin or MCP endpoint.
Configuration
When you run /plugin install, Claude Code prompts for that plugin's credentials (declared in each plugin's userConfig) and stores sensitive values in your system keychain. You can also set them via the env block in ~/.claude/settings.local.json, or export them in your shell profile.
better-notion-mcp, better-email-mcp, and better-workspace-mcp require credentials for their upstream APIs. better-telegram-mcp needs a bot token only for bot mode. Wet, Mnemo, and CRG preserve local Fastretrieval-based retrieval without provider API keys. Wet web search additionally needs a runnable backend: credential-free DuckDuckGo/Startpage, an external SearXNG endpoint, a configured cloud provider, or a local installation that can run SearXNG.
Cloud providers (wet-mcp, mnemo-mcp, better-code-review-graph)
Cloud model selection is separate from installing a plugin. Configure the model fields and matching provider keys in the server's environment for stdio or in the authenticated subject's relay form for multi-user HTTP. Do not infer a fallback chain just because several keys are present.
- Wet and Mnemo use
EMBEDDING_MODELS,RERANK_MODELS, andLLM_MODELS. - CRG uses
EMBEDDING_MODELSandSUMMARY_MODELS; its optional query reranker is selected byLOCAL_RERANK_MODEL, not a cloud rerank chain. - Empty/local configuration and cloud behavior are server-specific; consult the setup references.
The managed Cloudflare route uses exactly
openrouter/minimax/minimax-m3:free for completion and
cohere/embed-v4.0 / cohere/rerank-v4.0-fast for supported embedding/rerank
paths through Cloudflare AI Gateway. Cohere is paid and requires explicit
spending authorization with a cap before calls. There is no Jina, GLM, or paid
completion fallback in that route. CRG does not currently implement cloud
reranking. This server-side policy does not change user-owned OMP model/profile
settings or remove public local capabilities.
imagine-mcp
XAI_API_KEY=xai-... # default provider
GEMINI_API_KEY=AIza... # optional alternative
GOOGLE_VERTEX_EXPRESS_API_KEY=AQ... # optional alternative
OPENAI_API_KEY=sk-... # optional alternative
Telegram
TELEGRAM_BOT_TOKEN=123456:ABC-DEF...
Required for bot mode (from @BotFather); leave empty for user mode.
EMAIL_CREDENTIALS=user@gmail.com:app-password
Multiple accounts: user1@gmail.com:pass1,user2@outlook.com:pass2
Notion
NOTION_TOKEN=ntn_xxx
Get your token from notion.so/my-integrations.
Godot
GODOT_PATH=/path/to/godot # optional; auto-detects from PATH if empty
GODOT_PROJECT_PATH=/path/to/project # optional default project root
Google Workspace
GOOGLE_OAUTH_CLIENT_ID=<your-client-id>.apps.googleusercontent.com
GOOGLE_OAUTH_CLIENT_SECRET=<your-client-secret>
Create the OAuth client yourself at console.cloud.google.com/apis/credentials — type Desktop app for the default stdio mode, Web application for self-hosted HTTP. The first run opens Google's consent screen once, then reuses the stored refresh token.
Skills
Most plugins ship skills, invoked in Claude Code as <plugin>:<skill>:
| Plugin | Skills |
|---|---|
| wet-mcp | fact-check, compare, research-topic, lock-project-stack |
| mnemo-mcp | session-handoff, knowledge-audit, recall-context, memory-commit, passport-bootstrap |
| better-notion-mcp | organize-database, bulk-update |
| better-email-mcp | inbox-review, follow-up |
| better-telegram-mcp | setup-bot, channel-post |
| better-godot-mcp | build-scene, debug-issue, add-mechanic |
| better-code-review-graph | review-pr, review-delta, refactor-check |
| imagine-mcp | image-describe |
Other MCP Clients
Each plugin works with any MCP client. Run directly:
Python plugins (uvx / pipx / docker):
uvx --python 3.13 wet-mcp
uvx --python 3.13 mnemo-mcp
uvx --python 3.13 better-telegram-mcp
uvx --python 3.13 better-code-review-graph
uvx --python 3.13 imagine-mcp
TypeScript plugins (npx / bunx / docker):
npx -y @n24q02m/better-notion-mcp
npx -y @n24q02m/better-email-mcp
npx -y @n24q02m/better-godot-mcp
Add to any MCP client's settings.json:
{
"mcpServers": {
"wet": {
"command": "uvx",
"args": ["--python", "3.13", "wet-mcp"]
}
}
}
Deployment
The docs site (mcp.n24q02m.com) lives in this repo (docs/, Astro). Every push to main that changes docs/** or a plugin doc source (plugins/*/*.md, plugin manifests) triggers .github/workflows/deploy-docs.yml, which builds the site with Bun and deploys docs/dist to the Cloudflare Pages project mcp-n24q02m. Pages under docs/src/content/docs/servers/ are generated at build time by docs/scripts/sync-plugin-docs.mjs from the plugins/<name>/*.md sources — edit those source files, never the generated tree, and do not deploy the site manually with wrangler.
Plugin assets are synced separately: when a server repo cuts a release, cd.yml (plugin-sync) copies that release tag's plugin.json, skills, hooks, and commands into plugins/<name>/ on main, and the docs site redeploys if doc sources changed.
Contributing
See CONTRIBUTING.md.
License
Apache-2.0