Xpersona
agentCLAWHUBUnverified

智能文档助手

智能文档处理专家，支持文档解析、信息抽取、文档分类 Skill: 智能文档助手 Owner: ankylala Summary: 智能文档处理专家，支持文档解析、信息抽取、文档分类 Tags: latest:2.0.4 Version history: v2.0.4 | 2026-04-24T16:11:18.455Z | user - Version bump from 2.0.0 to 2.0.4 - No file or documentation changes detected in this release v2.0.3 | 2026-04-24T15:50:28.782Z | user - Enhanced documentation describing DocPilot’s high-precision document parsing, extraction, and classification capabilities. - Detailed explan

OpenClaw
View on ClawHubHomepage

Rank

62

Safety

84

Downloads

1.1k

Updated

Oct 11, 2026

Version

2.0.4

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 1.1K downloads reported by the source. Last updated 10/11/2026.

Avoid when

  • Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor
Clawhubvendor · observed Oct 11, 2026
Protocol compatibility
OpenClawcompatibility · observed Oct 11, 2026
Adoption signal
1.1K downloadsadoption · observed Oct 11, 2026
Latest release
2.0.4release · observed Apr 24, 2026
Handshake status
UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s1749k021s8ehj82qtdwad7emn841bhy:docpilot
  1. Setup complexity is LOW. This package is likely designed for quick installation with minimal external side-effects.
  2. Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-ankylala-docpilot/snapshot"

Documentation

CLAWHUB

48,066 characters of source documentation, loaded on request.

Extracted files

5 files captured from the source.

SKILL.md

---
name: DocPilot
description: 智能文档处理专家，支持文档解析、信息抽取、文档分类
version: 2.0.0
author: TokenAI
tags:
  - pdf
  - document
  - extraction
  - ocr
  - table
  - parser
  - classify
commands:
  - parse
  - extract
  - classify
---

# DocPilot — 智能文档处理专家

高精度文档处理技能，支持文档解析、信息抽取、文档分类。

## 为什么选择 DocPilot？

### 三层能力 + 六大核心优势

**三层能力**
1. **解析** — 高精度识别文档内容，保留版面结构
2. **抽取** — 按需求提取关键字段，每条结果都能溯源到原文位置
3. **分类** — 自动识别文档类型，混合文档也能自动切分

**六大核心优势**

#### 1. 证据溯源 — 每个字段都有"身份证" ⭐ 独家
```json
{
  "key": "合同金额",
  "value": "¥1,200,000",
  "confidence": "high",
  "evidence": [{
    "text": "合同总金额：¥1,200,000",
    "page": 2,
    "quad": [[120, 350], [480, 350], [480, 380], [120, 380]]
  }]
}
```
> 审计、法务、财务场景必备 — 知道数据从哪来，才能相信数据是对的。

#### 2. 混合文档切分 — 一份文件，多种类型 ⭐ 独家
上传一份包含"合同+发票+报价单"的混合文件，自动识别边界并逐段分类。

#### 3. 印章检测 — 公章/签名章/骑缝章自动识别 ⭐ 独家
自动检测文档中的印章和签章，返回位置和类型信息，适用于合同审查、资质验证。

#### 4. 跨页表格合并 — 断裂表格智能还原 ⭐ 独家
自动识别跨页断裂的表格，智能合并表头和表体，输出完整结构。

#### 5. 手写字体识别 — 印刷+手写混合识别 ⭐ 独家
支持印刷体和手写体混合识别，覆盖表单填写、手写批注、签字确认等场景。

#### 6. 全格式支持 — 一个技能全部搞定
PDF · 图片 · Word · Excel · CSV — 无需组合多个工具。

---

## 命令

### 解析文档
```
DocPilot parse <文件路径> [选项]
```

示例：
```
DocPilot parse C:\docs\report.pdf
DocPilot parse C:\docs\scan.jpg --output markdown
DocPilot parse C:\docs\data.xlsx
DocPilot parse C:\docs\contract.pdf --seal --bbox
```

### 信息抽取
```
DocPilot extract <文件路径> --schema <JSON>
```

示例：
```
DocPilot extract C:\docs\contract.pdf --schema "{\"fields\":[{\"key\":\"甲方\",\"type\":\"string\"},{\"key\":\"乙方\",\"type\":\"string\"}]}"
DocPilot extract C:\docs\invoice.pdf --schema schema.json
```

### 文档分类
```
DocPilot classify <文件路径> [选项]
```

示例：
```
DocPilot classify C:\docs\mixed.pdf
DocPilot classify C:\docs\docs.pdf --mode classify_and_split --categories "[{\"name\":\"合同\",\"description\":\"合同协议\"},{\"name\":\"发票\",\"description\":\"发票单据\"}]"
```

---

## 参数说明

### parse 命令

| 参数 | 说明 | 示例 |
|------|------|------|
| 文件路径 | PDF/图片/Word/Excel 文件路径 | `C:\docs\report.pdf` |
| --output | 输出格式 (structured/markdown/text) | `--output markdown` |
| --layout | 启用版面分析 | `--layout` |
| --table | 启用表格识别（含跨页合并） | `--table` |
| --seal | 启用印章识别 | `--seal` |
| --dpi | DPI (72/144/200/216) | `--dpi 200` |
| --pages | 页码范围 | `--pages 1-5,8,10-12` |
| --bbox | 包含边界框坐标 | `--bbox` |
| --normalize | 返回格式化解析数据 (默认开启) | `--normalize` |
| --raw | 返回原始解析格式 | `--raw` |
| --include-image | markdown 中包含图片 | `--include-image` |
| --image-format | 图片格式 (url/base64) | `--image-format url` |

### extract 命令

| 参数 | 说明 | 示例 |
|------|------|------|
| 文件路径 | 文档文件路径 | `C:\docs\contract.pdf` |
| --schema | 字段 schema（必填） | `--schema '{"fields":[...]}'` |
| --prompt | 提示词模式 schema | `--prompt '{"fields":[...]}'` |
| --schema-ref | 模板引用 | `--schema-ref DocPilot/contract/v1` |
| --options | 扩展配置 | `--options '{"mode":"fast"}'` |

### classify 命令

| 参数 | 说明 | 示例 |
|------|------|------|
| 文件路径 | 文档文件路径 | `C:\docs\mixed.pdf` |
| --mode | 分类模式 | `--mode classify_and_split` |
| --categories | 分类 schema | `--categories '[{"name":"合同","description":"..

README.md

# DocPilot — Intelligent Document Processing Expert

> OpenClaw Skill - Document Parsing, Information Extraction, and Classification

Extract structured data from PDF, images, Word, and Excel documents. Perform field extraction and document classification.

## Why Choose DocPilot?

### Three-Layer Capabilities + Six Core Advantages

**Three-Layer Capabilities**
1. **Parse** — High-precision document content recognition with layout preservation
2. **Extract** — Extract key fields with evidence tracing back to original positions
3. **Classify** — Auto-identify document types, split mixed documents automatically

**Six Core Advantages**

#### 1. Evidence Tracing — Every Field Has an "ID Card" ⭐ Exclusive
```json
{
  "key": "Contract Amount",
  "value": "¥1,200,000",
  "confidence": "high",
  "evidence": [{
    "text": "Total Contract Amount: ¥1,200,000",
    "page": 2,
    "quad": [[120, 350], [480, 350], [480, 380], [120, 380]]
  }]
}
```
> Essential for audit, legal, and finance — know where data comes from to trust it.

#### 2. Mixed Document Splitting — One File, Multiple Types ⭐ Exclusive
Upload a mixed file containing "contract + invoice + quotation", auto-identify boundaries and classify segment by segment.

#### 3. Seal Detection — Official/Signature/Cross-page Seals ⭐ Exclusive
Auto-detect seals and stamps in documents, return position and type info. Ideal for contract review and qualification verification.

#### 4. Cross-page Table Merging — Smart Reconstruction ⭐ Exclusive
Auto-identify tables split across pages, intelligently merge headers and bodies, output complete structures.

#### 5. Handwriting Recognition — Print + Handwriting Mixed ⭐ Exclusive
Support mixed recognition of printed and handwritten text. Covers form filling, handwritten annotations, and signature confirmation.

#### 6. Full Format Support — One Skill Does It All
PDF · Images · Word · Excel · CSV — No need to combine multiple tools.

---

## Features

- ✅ **Multi-format Support**: PDF, Images (JPG/PNG), Word, Excel, CSV
- ✅ **Layout Analysis**: Automatically detect and structure document elements
- ✅ **Table Recognition**: Extract tables with HTML and Markdown outputs, cross-page merge
- ✅ **Information Extraction**: Extract structured fields with evidence tracing
- ✅ **Document Classification**: Single document or mixed document classification & splitting
- ✅ **Seal Detection**: Detect stamps and seals in documents
- ✅ **Handwriting Recognition**: Print + handwriting mixed recognition
- ✅ **Multiple Output Formats**: structured JSON, markdown, plain text

---

## Quick Start

### Installation

```bash
# Install via ClawHub
openclaw skills install DocPilot

# Or manual installation (local development)
cd E:\skills\document-parser
pip install -r requirements.txt
```

### Configuration

**Option 1: Environment Variables**
```bash
# Windows PowerShell
$env:DOCPilot_BASE_URL="https://docpilot.token-ai.com.cn"
$env:DOCPilot_API_KEY="your_api_key"
```

**Option 2: Configura

_meta.json

{
  "ownerId": "kn7ewjb6a59zsr2zz60ky1wrzx82gdqg",
  "slug": "docpilot",
  "version": "2.0.4",
  "publishedAt": 1777047078455
}

README_zh.md

# DocPilot — 智能文档处理专家

> OpenClaw Skill - 高精度文档解析、信息抽取、文档分类

从 PDF、图片、Word、Excel 文档中提取结构化数据，支持信息抽取和文档分类。

## 为什么选择 DocPilot？

### 三层能力 + 六大核心优势

**三层能力**
1. **解析** — 高精度识别文档内容，保留版面结构
2. **抽取** — 按需求提取关键字段，每条结果都能溯源到原文位置
3. **分类** — 自动识别文档类型，混合文档也能自动切分

**六大核心优势**

#### 1. 证据溯源 — 每个字段都有"身份证" ⭐ 独家
```json
{
  "key": "合同金额",
  "value": "¥1,200,000",
  "confidence": "high",
  "evidence": [{
    "text": "合同总金额：¥1,200,000",
    "page": 2,
    "quad": [[120, 350], [480, 350], [480, 380], [120, 380]]
  }]
}
```
> 审计、法务、财务场景必备 — 知道数据从哪来，才能相信数据是对的。

#### 2. 混合文档切分 — 一份文件，多种类型 ⭐ 独家
上传一份包含"合同+发票+报价单"的混合文件，自动识别边界并逐段分类。

#### 3. 印章检测 — 公章/签名章/骑缝章自动识别 ⭐ 独家
自动检测文档中的印章和签章，返回位置和类型信息，适用于合同审查、资质验证。

#### 4. 跨页表格合并 — 断裂表格智能还原 ⭐ 独家
自动识别跨页断裂的表格，智能合并表头和表体，输出完整结构。

#### 5. 手写字体识别 — 印刷+手写混合识别 ⭐ 独家
支持印刷体和手写体混合识别，覆盖表单填写、手写批注、签字确认等场景。

#### 6. 全格式支持 — 一个技能全部搞定
PDF · 图片 · Word · Excel · CSV — 无需组合多个工具。

---

## 功能特性

- ✅ **多格式支持**: PDF、图片 (JPG/PNG)、Word、Excel、CSV
- ✅ **版面分析**: 自动检测和解析文档结构元素
- ✅ **表格识别**: 提取表格并输出 HTML 和 Markdown 格式，支持跨页合并
- ✅ **信息抽取**: 按 schema 提取结构化字段，带证据溯源
- ✅ **文档分类**: 单文档分类或混合文档切分分类
- ✅ **印章检测**: 检测文档中的印章和签章
- ✅ **手写识别**: 印刷体+手写体混合识别
- ✅ **多种输出格式**: structured JSON、markdown、纯文本

---

## 快速开始

### 安装

```bash
# 通过 ClawHub 安装
openclaw skills install DocPilot

# 或手动安装（本地开发）
cd E:\skills\document-parser
pip install -r requirements.txt
```

### 配置

**方式一：环境变量**
```bash
# Windows PowerShell
$env:DOCPilot_BASE_URL="https://docpilot.token-ai.com.cn"
$env:DOCPilot_API_KEY="your_api_key"
```

**方式二：配置文件**
```bash
cd E:\skills\DocPilot
copy config.example.json config.json
# 编辑 config.json 填入你的 API 地址和 API Key
```

**方式二：配置文件**
```bash
cd E:\skills\document-parser
copy config.example.json config.json
# 编辑 config.json 填入你的 API 地址
```

### 使用方法

#### 解析文档
```bash
# 基础解析
DocPilot parse "C:\docs\report.pdf"

# Markdown 输出
DocPilot parse "C:\docs\scan.jpg" --output markdown

# 启用印章检测和边界框
DocPilot parse "C:\docs\contract.pdf" --seal --bbox

# 解析 Excel
DocPilot parse "C:\docs\data.xlsx"
```

#### 信息抽取
```bash
# 按 schema 抽取字段
DocPilot extract "C:\docs\contract.pdf" --schema "{\"fields\":[{\"key\":\"甲方\",\"type\":\"string\"},{\"key\":\"乙方\",\"type\":\"string\"}]}"

# 使用 schema 文件
DocPilot extract "C:\docs\invoice.pdf" --schema schema.json
```

#### 文档分类
```bash
# 单文档分类
DocPilot classify "C:\docs\doc.pdf"

# 混合文档分类和切分
DocPilot classify "C:\docs\mixed.pdf" --mode classify_and_split --categories "[{\"name\":\"合同\",\"description\":\"合同协议\"},{\"name\":\"发票\",\"description\":\"发票单据\"}]"
```

---

## 参数说明

### parse 命令

| 参数 | 类型 | 必填 | 说明 |
|------|------|------|------|
| file | string | 是 | PDF/图片/Word/Excel 文件路径 |
| --output | string | 否 | 输出格式：structured/markdown/text |
| --layout | flag | 否 | 启用版面分析 |
| --table | flag | 否 | 启用表格识别（含跨页合并） |
| --seal | flag | 否 | 启用印章检测 |
| --dpi | int | 否 | DPI：72/144/200/216 |
| --pages | string | 否 | 页码范围，如 "1-5,8,10-12" |
| --bbox | flag | 否 | 包含边界框坐标 |
| --normalize | flag | 否 | 返回格式化解析数据（默认开启） |
| --raw | flag | 否 | 返回原始解析格式 |
| --include-ima

skill-card.md

## Description:

智能文档处理专家，支持文档解析、信息抽取、文档分类。

This skill is ready for commercial/non-commercial use.

## Publisher:

[ankylala](https://clawhub.ai/user/ankylala)

### License/Terms of Use:

MIT-0

## Use Case:

External users and developers use this skill to parse PDFs, images, Word, Excel, and CSV files, extract structured fields with source evidence, and classify or split mixed document sets. Typical workflows include contract review, finance audits, archive organization, form processing, and compliance checks.

### Deployment Geography for Use:

Global

## Known Risks and Mitigations:

Risk: The skill uploads selected documents to the DocPilot/TokenAI service or another configured endpoint, and privacy controls are under-documented.

Mitigation: Use only documents approved for that external service until the publisher documents retention, deletion, jurisdiction, and privacy terms.

Risk: The bundled configuration includes credential-like data and can override environment variables.

Mitigation: Remove bundled secrets, rotate any exposed credentials, and require deployers to provide API keys through approved secret management.

Risk: The service destination is configurable, which can send document contents to an unexpected endpoint.

Mitigation: Restrict deployment to approved HTTPS destinations and review endpoint configuration before each release.

Risk: Dependencies are not pinned.

Mitigation: Pin and review dependencies before production deployment.

## Reference(s):

- [ClawHub Skill Page](https://clawhub.ai/ankylala/skills/docpilot)
- [DocPilot Service Endpoint](https://docpilot.token-ai.com.cn)

## Skill Output:

**Output Type(s):** [text, markdown, configuration, guidance]

**Output Format:** [JSON, Markdown, or plain text document-processing results]

**Output Parameters:** [1D]

**Other Properties Related to Output:** [Parse outputs may include document IDs, page counts, layout elements, tables, seals, bounding boxes, and markdown; extraction and classification outputs include structured fields, evidence, segments, confidence, and metadata.]

## Skill Version(s):

2.0.4 (source: server release evidence; source files report 2.0.0)

## Ethical Considerations:

Users should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.

Related agents

Same protocol, similar job.

Github ReposUpdated 2d agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW
Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW
Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.

MCPOPENCLAW
Github ReposUpdated 7mo agoRank 70

CopilotKit

The Frontend for Agents & Generative UI. React + Angular

OPENCLAW

Machine-readable data

The same record, as JSON, for agents and crawlers.

{
  "facts": [
    {
      "factKey": "vendor",
      "category": "vendor",
      "label": "Vendor",
      "value": "Clawhub",
      "href": "https://clawhub.ai/ankylala/skills/docpilot",
      "sourceUrl": "https://clawhub.ai/ankylala/skills/docpilot",
      "sourceType": "profile",
      "confidence": "medium",
      "observedAt": "2026-10-11T13:05:41.105Z",
      "isPublic": true
    },
    {
      "factKey": "protocols",
      "category": "compatibility",
      "label": "Protocol compatibility",
      "value": "OpenClaw",
      "href": "https://www.xpersona.co/api/v1/agents/clawhub-ankylala-docpilot/contract",
      "sourceUrl": "https://www.xpersona.co/api/v1/agents/clawhub-ankylala-docpilot/contract",
      "sourceType": "contract",
      "confidence": "medium",
      "observedAt": "2026-10-11T13:05:41.105Z",
      "isPublic": true
    },
    {
      "factKey": "traction",
      "category": "adoption",
      "label": "Adoption signal",
      "value": "1.1K downloads",
      "href": "https://clawhub.ai/ankylala/docpilot",
      "sourceUrl": "https://clawhub.ai/ankylala/docpilot",
      "sourceType": "profile",
      "confidence": "medium",
      "observedAt": "2026-10-11T13:05:41.105Z",
      "isPublic": true
    },
    {
      "factKey": "latest_release",
      "category": "release",
      "label": "Latest release",
      "value": "2.0.4",
      "href": "https://clawhub.ai/ankylala/docpilot",
      "sourceUrl": "https://clawhub.ai/ankylala/docpilot",
      "sourceType": "release",
      "confidence": "medium",
      "observedAt": "2026-04-24T16:11:18.455Z",
      "isPublic": true
    },
    {
      "factKey": "handshake_status",
      "category": "security",
      "label": "Handshake status",
      "value": "UNKNOWN",
      "href": "https://www.xpersona.co/api/v1/agents/clawhub-ankylala-docpilot/trust",
      "sourceUrl": "https://www.xpersona.co/api/v1/agents/clawhub-ankylala-docpilot/trust",
      "sourceType": "trust",
      "confidence": "medium",
      "observedAt": null,
      "isPublic": true
    }
  ],
  "events": [
    {
      "eventType": "release",
      "title": "Release 2.0.4",
      "description": "- Version bump from 2.0.0 to 2.0.4 - No file or documentation changes detected in this release",
      "href": "https://clawhub.ai/ankylala/docpilot",
      "sourceUrl": "https://clawhub.ai/ankylala/docpilot",
      "sourceType": "release",
      "confidence": "medium",
      "observedAt": "2026-04-24T16:11:18.455Z",
      "isPublic": true
    }
  ]
}

Record generated Oct 11, 2026.

Sponsored

Ads related to 智能文档助手 and adjacent AI workflows.