跳过安装。2 分钟内即可运行。
在 cloud.anythingmcp.com 开始免费试用，一键添加 Paperless-ngx，然后将你的 AI 客户端（Claude、ChatGPT、Copilot 或 Cursor）指向生成的 MCP 端点。无需 Docker，无需 git clone，无需任何工程经验。
Summary
Paperless-ngx document management: full-text search over scanned documents, correspondents, document types, tags, storage paths and the OCR'd content of any document. REST API on your own instance, token auth.
试试这样问
Paperless-ngx 的示例提示
点击任意提示即可复制，然后粘贴到 Claude、ChatGPT、Cursor、Gemini、Copilot 或 OpenClaw 中，在该连接器上运行。
Claude 是 AI，可能会出错。请核对回答。
💡 无需安装? 直接使用 cloud.anythingmcp.com。 登录、点击 Connectors → Paperless-ngx、粘贴凭证、生成 MCP API key — 完成。无 Docker、无
git clone、无本地服务器。
Paperless-ngx + Claude
Paperless-ngx document management: full-text search over scanned documents, correspondents, document types, tags, storage paths and the OCR'd content of any document. REST API on your own instance, token auth.
前提条件
完整的设置说明已内置在连接器中 (在 store 中选择连接器时可见)。所需环境变量:
PAPERLESS_URL, PAPERLESS_API_TOKEN
步骤 1 — 获取凭证
Getting a token
- Sign in to Paperless-ngx, open the user menu → My Profile, and copy the API Auth Token (create one if the field is empty).
- Put it in
PAPERLESS_API_TOKENand your instance's base URL inPAPERLESS_URL— no trailing slash and no/api, e.g.https://paperless.example.com.
The Authorization header is Token …, Django REST Framework's scheme — not Bearer. The adapter sends it correctly.
…(continued in the in-app connector instructions)
步骤 2 — 安装 adapter
curl -fsSL https://raw.githubusercontent.com/HelpCode-ai/anythingmcp/main/docker-compose.quickstart.yml -o docker-compose.yml
printf 'JWT_SECRET=%s\nENCRYPTION_KEY=%s\n' "$(openssl rand -hex 32)" "$(openssl rand -hex 32)" > .env
docker compose up -d
步骤 3 — 在 Claude 中添加连接器
推荐路径 — 在 claude.ai web 上工作，无需触碰任何配置文件。
- 打开 claude.ai/customize/connectors。
- 点击 "Add custom connector"。
- 填入:
- Name:
Paperless-ngx - URL:
https://cloud.anythingmcp.com/mcp - Authentication: Bearer token → 粘贴你的 MCP API key (来自 AnythingMCP → Profile → MCP API Keys → New Key)
- Name:
- 点击 Connect 授权。
下面所有工具都会出现在你的聊天中 — 开始输入提示。
可用工具
| Tool | What it does |
|---|---|
paperless_ngx_search_documents | Full-text search across the OCR'd content of every document |
paperless_ngx_list_documents | List documents filtered by correspondent, document type, tag or creation date — the structured alternative to full-text search when you alre |
paperless_ngx_get_document | Read one document's metadata: title, correspondent, type, tags, created and added dates, archive serial number and the full OCR'd content |
paperless_ngx_get_document_metadata | Read the technical metadata of one document: original filename, media type, checksum, page count and any embedded PDF metadata |
paperless_ngx_list_correspondents | List correspondents — the senders documents are filed under — with their document count and matching rules |
paperless_ngx_list_document_types | List the document types configured in the instance (invoice, contract, payslip …) with their document counts |
paperless_ngx_list_tags | List tags with their colour, inbox flag and document count |
paperless_ngx_list_storage_paths | List the storage paths documents are filed into, with their path template and document count |
paperless_ngx_get_statistics | Read instance-wide statistics: total documents, documents in the inbox, character count and the breakdown by file type |
FAQ
Claude Code 和 Claude Desktop 一样工作吗? 是的 — 两者都指向 https://cloud.anythingmcp.com/mcp。
下一步
这份指南对你有帮助吗?
