Office Document Operations
SkillDocs & knowledgeChoose the right extraction or editing path for office documents such as PDF and DOCX without losing review context.
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 Office Document Operations skill
What this skill tells your AI
The instructions your AI receives, as published by contextgo/contextgo in src/process/resources/skills/office-analyst-pack/skills/office-document-operations/SKILL.md and read by ahel’s review.
Use this skill when office documents need more than a quick skim.
What this skill does
It helps choose the right path for reading, extracting, reviewing, or editing office documents.
Use when
- PDFs or DOCX files are central to the task.
- The user needs table extraction, clean summaries, tracked review, or structured document changes.
- Multiple office documents need to be compared or synthesized.
Do not use when
- The work is fundamentally spreadsheet-first.
- The user only needs a casual summary and no document fidelity matters.
Decision tree
Reading mode
Use when you need:
- a structured summary
- key clauses or decisions
- table extraction
- issue spotting
Review mode
Use when the document is formal and changes should preserve review context:
- business memos
- contracts
- policy docs
- external-facing drafts
Default to a tracked or review-aware path instead of silent overwrites.
Clean-output mode
Use when the user wants a polished final version with no visible review scaffolding.
Workflow
Step 1: Classify each file
For each document, identify:
- format: PDF or DOCX
- likely role: source evidence, draft for editing, final artifact, or appendix
- whether tables, comments, tracked changes, or form fields matter
Step 2: Pick the extraction path
Examples:
- PDF with text and tables -> extract text and tables separately
- DOCX with business content -> preserve structure and review context
- DOCX draft for revision -> choose review-aware editing if external or formal
Step 3: Preserve what matters
Do not accidentally erase:
- tracked changes
- comments
- section structure
- table layout that carries meaning
Step 4: Convert extraction into a usable office output
Turn the raw extraction into:
- summary
- clause list
- action list
- issue register
- decision memo input
Output format
Return:
1. File map
- each file
- role
- chosen processing path
2. Extracted signal
- main points
- important tables or structured content
- review context that should be preserved
3. Risks
- ambiguity
- missing pages or broken extraction
- review fidelity concerns
Use together with
pdfdocxoffice-briefing
Signals
- GitHub stars
- 54
- Forks
- 5
- Last commit
- May 2026
Advanced
- Catalog kind
- skill
- Gateway key
office-document-operations- Source
- github.com/contextgo/contextgo