# Office Docs

> Office Docs is an Agent Skill for documents published in the RTGS2017/NagaAgent repository on GitHub: a SKILL.md instruction file an AI agent can load to do this task.

Extract text and tables from .docx and .xlsx using local scripts (no external deps).

- Skill name: `office-docs`
- Category: [Documents](https://www.openparable.com/skill/documents/1)
- Source: https://github.com/RTGS2017/NagaAgent/blob/HEAD/apiserver/skills_templates/office-docs/SKILL.md
- Repository: RTGS2017/NagaAgent (1,538 GitHub stars when collected)

## SKILL.md

# Office Docs (docx + xlsx)

This skill provides lightweight, dependency-free extraction for Office documents using the scripts in `tools/`.

## DOCX

Extract paragraphs and tables from a `.docx` file:

```bash
python tools/docx_extract.py "/path/to/file.docx" --output "/path/to/output.txt"
```

## XLSX

Extract sheet data from a `.xlsx` file:

```bash
python tools/xlsx_extract.py "/path/to/file.xlsx" --output "/path/to/output.txt" --format csv
```

### Notes

- If `--output` points to a directory, each sheet is written to its own file.
- If `--output` is a file and there are multiple sheets, the script prefixes each sheet with `# Sheet: <name>`.

---

Source: https://www.openparable.com/skill/documents/office-docs (OpenParable). Cite the canonical URL above.
