multiplist-mcp-server
Multiplist is an IP vault with citations, acting as an interoperable source of truth for agents. It extracts decisions, frameworks, and key passages from docs and AI conversations with character‑level provenance, so every claim traces back to exactly what you said—across 19 curated chat‑native tools in six categories: Capture, Extraction, Custom Skills, Discovery, Research, and Annotation.
README
@multiplist/mcp-server
IP Vault with Citations — extract decisions, frameworks, and insights from your AI conversations with full provenance. Every claim traces back to exactly what you said.
Your AI conversations are intellectual property. Multiplist treats them that way.
Connect via MCP
Multiplist works as an MCP server. Connect from Claude Desktop, Claude Code, or any MCP-compatible client.
Endpoint: https://multiplist.ai/mcp
Transport: StreamableHTTP + OAuth 2.1 with PKCE
Claude Desktop
Add to your claude_desktop_config.json:
{
"mcpServers": {
"multiplist": {
"command": "npx",
"args": ["-y", "@multiplist/mcp-server"],
"env": {
"MULTIPLIST_API_KEY": "your-key-here"
}
}
}
}
Or connect directly via the hosted endpoint:
{
"mcpServers": {
"multiplist": {
"type": "streamable-http",
"url": "https://multiplist.ai/mcp"
}
}
}
19 Curated Tools
| Category | Tools | What They Do |
|---|---|---|
| Capture | create_source, update_source, delete_source | Save conversations to your vault |
| Extraction | trigger_extraction, batch_extract, batch_status | Extract structured meaning |
| Custom Skills | create_skill, list_skills, get_skill, match_skills | Define domain-specific extraction |
| Discovery | get_source, search_vault, search_sources, get_vault_summary, get_vault_gaps | Search and retrieve with citations |
| Research | request_brief, get_brief_status | Synthesize across your vault |
| Annotate | add_annotation, add_marginalia | Write corrections and evolution notes |
What Multiplist Extracts
Nine categories of structured meaning from any conversation or document:
- Decisions — commitments, with reasoning
- Frameworks — named models and structured approaches
- Actions — next steps, with ownership
- Questions — open threads and unresolved tensions
- Definitions — terms with precise meaning in context
- Emergence — new concepts forming in real time
- Golden Passages — language worth preserving verbatim
- Exemplars — concrete examples worth referencing later
- Offers — AI suggestions and proposed next steps
Each seed traces back to exact character positions in the original source.
Custom Extraction Skills
The differentiator: you define what matters in your domain.
A coach creates a "Client Breakthroughs" skill. A developer creates an "Architecture Signals" skill. A researcher creates a "Research Distillation" skill.
Skills tell Multiplist what to look for in your specific content type. The vault learns the shape of your work.
Get Started
- Sign up at multiplist.ai
- Generate an API key in Settings
- Connect via MCP (see above)
- Say "save this conversation to my vault"
Skills Library
See github.com/multiplist/skills for pre-built skills and the Seed Doc Renderer, Archivist Research, and Custom Extraction Skills integrations.
License
MIT — multiplist.ai
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。