Office Workflows
SkillFiles & storageCoordinate cross-format office workflows across CSV/TSV, XLSX, DOCX, and PPTX deliverables.
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 Workflows skill
What this skill tells your AI
The instructions your AI receives, as published by hybridaione/hybridclaw in skills/office-workflows/SKILL.md and read by ahel’s review.
Use this skill when the task spans multiple office formats or involves both analysis and polished deliverables.
Fresh Outputs By Default
- When the user asks for a new deliverable from a folder of source files, build it fresh from the primary inputs named by the task.
- Do not inspect or reuse older generated artifacts in that folder such as dashboards, summary files, helper scripts, or prior reports unless the user explicitly asks to update them or use them as templates.
Recommended Patterns
- CSV/TSV -> XLSX:
Run
node skills/xlsx/scripts/import_delimited.cjs input.csv output.xlsx --json, validate the workbook, then polish the structure with thexlsxskill. - XLSX -> PPTX summary:
Use
delegateinchainmode. Step 1: inspect the workbook and extract the few metrics or charts that matter. Step 2: build the deck with thepptxskill. Step 3: check whethersoffice/libreofficeandpdftoppmexist:sh -lc 'command -v soffice >/dev/null 2>&1 || command -v libreoffice >/dev/null 2>&1'sh -lc 'command -v pdftoppm >/dev/null 2>&1'Step 4: only if both checks succeed, run the PPTX render-and-review loop on thumbnails until there are no concrete slide-level issues left. Otherwise skip that QA path silently unless the user explicitly asked for QA, export, thumbnails, validation, or render verification. - XLSX -> DOCX memo:
Analyze the workbook first, then draft the memo in Markdown and convert with
pandocor generate/edit a DOCX directly.
Delegation Guidance
- Use
parallelmode for independent research branches, such as "analyze workbook tabs" and "inspect brand template assets". - Use
chainmode when later outputs depend on prior findings. - Keep each delegated prompt explicit about file paths, expected output, and whether the subagent should modify files or only return findings.
Templates
- Prefer user-supplied templates from current-turn attachments or mounted workspace paths.
- For branded deliverables, inspect the template first, then preserve its structure while swapping in updated data and copy.
Signals
- GitHub stars
- 134
- Forks
- 13
- Last commit
- Sep 2026
Advanced
- Catalog kind
- skill
- Gateway key
office-workflows- Source
- github.com/hybridaione/hybridclaw