多格式文档解析 Skill(密卷房)
SkillFiles & storageLets your agent turn PDF, Word, Excel, and PowerPoint documents into clean Markdown text with metadata and a confidence score.
Available today. Use it from your connected AI after setup.
No other account needed.
Connect ahel once, and every AI you use reads what you have installed.
Then ask your AI: use the 多格式文档解析 Skill(密卷房) skill
About this capability
Parses multi-format documents such as PDF, PPT, Excel, and Word into structured Markdown, outputting metadata and parsing confidence scores. Serves as the data foundation for RAG and the four-color cards.
What this skill tells your AI
The instructions your AI receives, as published by anbeime/skill in antinet-agentteams/skills/doc-parse/SKILL.md and read by ahel’s review.
使用方式
- 由密卷房 Worker 在收到已通过安全扫描的文件时调用。
- 三级 fallback 依次尝试,输出最终结构化结果与置信度。
输入(Input)
file_path:已通过 security-scan 的本地文件路径formats:(可选)期望支持的格式白名单,默认全格式
输出(Output)
markdown:结构化 Markdown 正文metadata:标题、页数、表格数、作者等元数据confidence:0–1 解析置信度fallback_used:最终生效的解析器名称
依赖(Dependencies)
- MinerU(首选,强排版还原)
- PyMuPDF(次选,PDF 快速解析)
- pdfplumber(兜底,表格/文本抽取)
python-magic(类型探测)
失败处理(Failure Handling)
- 主解析器失败 → 自动降级到下一档,直到全部尝试。
- 三级全部失败 → 标记
人工介入,不输出残缺结果,回传 BLOCKED 给军机处。 - 单页超大文件 → 分块解析后拼接,避免内存溢出;块级失败仅标记该块低置信度。
复用价值(Reuse Value)
- 通用解析底座:RAG 索引、企业知识库、合同结构化均可直接复用。
- 置信度透明:下游(通政司四色卡片)可据此决定是否需要人工复核,降低幻觉风险。
复赛代码包执行(runnable package)
- 真实入口:
scripts/run_doc_parse.py - 执行等价于
core.runtime.AgentSession.run_stage("doc-parse"),调用archive.mijuanfang.MiJuanFangAgent(三级解析 fallback,纯 Python 可离线)。 - 运行:
python skills/doc-parse/scripts/run_doc_parse.py - 产物:
examples/snse_survey/skill_outputs/doc_parse.json(解析结果 + 置信度 + fallback 信息)。
Signals
- GitHub stars
- 7k
- Forks
- 628
- Last commit
- Sep 2026
ahel recommends instead
Advanced
- Catalog kind
- skill
- Gateway key
doc-parse- Source
- github.com/anbeime/skill