zhr

What an agent can do with the docforge corpus

attnstack/docforge: 中文扫描件 / 复杂表格解析引擎 + 公开人工标注 benchmark

• Look up how docforge implements benchmark, chinese-nlp, document-parsing, with answers cited to file and line.
• Reuse a worked example: the commit “真实公文集 v0.1-rc + OCR Windows 原生验证 + 真实数据驱动的引擎改进” (16 files · Python).
• Answer from docforge's real Python source instead of memory (AI, prompts and agents).

• Where does docforge handle “真实公文集 v0.1-rc + OCR Windows 原生验证 + 真实数据驱动的引擎改进”?
• How does docforge implement benchmark?
• What does benchmark do in docforge?

Search the GitHub repository attnstack/docforge with XERJ (https://hub.xerj.org/corpus/attnstack--docforge).
Run these in a shell, in order, and check each result before the next:
1. xerj --help  (success: it prints usage). If xerj is missing: curl -fsSL https://xerj.org/get | sh
2. git clone https://github.com/attnstack/docforge docforge && git -C docforge checkout 62de89c5e81e1dc85b6f52d34663125d5d6da51c
3. Start a local node and leave it running: xerj --insecure --data-dir ./xerj-data
4. xerj autoindex ./docforge --url http://127.0.0.1:9200 --prefix docforge --progress plain
5. xerj autoindex map --url http://127.0.0.1:9200
Then answer my question from the indexed files and cite the ax_path of every hit. Query syntax: https://xerj.org/llms.txt
Licence: MIT. Keep the copyright notice and credit the authors.

corpus attnstack--docforge · AI, prompts and agents
Related corpora: oxidize-pdf, ann-benchmarks, docling-project/docling, lucasdmarshall/Hirara
AI search and retrieval with XERJ · Case study: reference coding for agents