GenOffice is an open-source office suite for editing Word, Excel, PowerPoint, PDF, and Markdown files with AI-assisted document editing. People use it as an alternative to Microsoft Office on macOS, Windows, and Linux.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add genspark-ai/genoffice --skill business-reportgit clone --depth 1 https://github.com/genspark-ai/genofficeWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/genspark-ai/genoffice/business-report)<a href="https://agentmods.dev/skills/genspark-ai/genoffice/business-report"><img src="https://agentmods.dev/badge/skills/genspark-ai/genoffice/business-report/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/genspark-ai/genoffice/business-report"><img src="https://agentmods.dev/badge/skills/genspark-ai/genoffice/business-report.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00060 | $0.00921 |
| Opus 5.5 | $0.00024 | $0.00368 |
| Sonnet 5.5 | $0.00012 | $0.00184 |
| Haiku 4.5 | $0.00006 | $0.00092 |
Grade A, and why
business-report scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 82 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Business report
You are drafting a report that a busy reader will skim first and read second. Structure carries the argument; prose fills it in.
Before writing
- Pin down four things from the request and the attached files. Ask only if one is genuinely missing and cannot be inferred: the decision the report supports, the reader (executive, team, client, regulator), the time period or scope, and the length (default: 2-4 pages).
- Read every attached or referenced file in full before drafting. Numbers, names and dates come from those sources, never from memory. If a source is silent, say so in the report rather than inventing a figure.
- Write the one-sentence conclusion first. If you cannot state it in one sentence, you are not ready to write.
Structure
Use these headings in this order, as real heading styles (Heading 1), not bold paragraphs. Drop a section only when the request rules it out.
- Executive summary — 3-6 sentences. The conclusion, the two or three facts that carry it, and what the reader is asked to decide or do. No background, no methodology.
- Background — Why the question exists now. Scope and period. What was examined and what was left out, in one short paragraph each.
- Findings (or Analysis) — One subsection (Heading 2) per finding, ordered by importance to the decision, not by the order you discovered them. Each subsection: a one-line finding as its first sentence, then the evidence. Put comparisons in a table; put trends in a short bullet list of figures, each with its period and source.
- Recommendations — Numbered. Each one is an action with an owner and a time frame when the sources allow it, followed by one sentence on the cost or risk of not doing it. Recommendations must trace back to a finding above; do not introduce new facts here.
- Risks and open questions — What could make the conclusion wrong, and what is still unknown. Short bullets.
- Appendix — Source list, data tables too large for the body, definitions. Only when there is something to put here.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 3d ago First seen · 82 lines · 60 tokens per session scan A 0fcbe726c512
business-report is a skill published in the GitHub repository genspark-ai/genoffice (10,833 stars, last pushed 2d ago), licensed Apache-2.0. It adds 60 tokens to every session and 921 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-23.
Other skills, from other repositories
office-router
Route Office requests into the correct xlsx, docx, or pptx workflow, while preferring Slidev for new presentation drafts unless .pptx is explicitly required.
geo-proposal
Auto-generate a professional, client-ready GEO service proposal from audit data. Creates a full proposal in markdown and PDF including executive summary, findings, recommended service packages (Basic/Standard/Premium), pricing, timeline, and terms. Use when user says "proposal", "proposta", "offerta", "preventivo"…
document-generation
Generate Word (.docx), Excel (.xlsx) and PowerPoint (.pptx) documents and fill existing PDF forms, from real NetClaw data, with per-element provenance and no fabrication. This is the generic document-writing engine (tool surface, tagged-value shapes, fabrication rules). For the four standard NetClaw report…
document-processing
Use when the deliverable is a document's bytes or its literal content — text/tables out of PDFs, AcroForm fill and flatten, page merge/split, PDF/DOCX from templates, OCR of image-only scans. NOT schema-typed fields pulled from text (that is structured-extraction), NOT signature routing (e-signature) or spreadsheet…
dingtalk-table
Read and manage DingTalk spreadsheets through the DingTalk Table MCP. Use this skill when the user explicitly refers to a DingTalk spreadsheet, worksheet, cell, or formula.
office-artifacts
Manage Office files, formats and versions; use Writer, Calc or Impress for content.