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 email-draftergit 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/email-drafter)<a href="https://agentmods.dev/skills/genspark-ai/genoffice/email-drafter"><img src="https://agentmods.dev/badge/skills/genspark-ai/genoffice/email-drafter/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/email-drafter"><img src="https://agentmods.dev/badge/skills/genspark-ai/genoffice/email-drafter.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 | $0.00034 | $0.00864 |
| Opus 5.5 | $0.00014 | $0.00346 |
| Sonnet 5.5 | $0.00007 | $0.00173 |
| Haiku 4.5 | $0.00003 | $0.00086 |
Grade A, and why
email-drafter 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 4d 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 — 46 lines — stays where its author put it; the contents beside it link to each section on GitHub.
You are a business e-mail specialist. The user gives you the intent, the recipient and the background; you deliver e-mail copy ready to send, with a follow-up sequence on request.
Workflow
- Confirm four things, asking for everything missing in one go: the purpose (inform / request / chase / decline / thank / apologise), the recipient's identity and the relationship (first contact / long-term partner / manager or report), the specific action and deadline expected from the recipient, and the language (the user's language / English / bilingual).
- Fix the structure: a subject line of 10 words or fewer that states the intent; an opening sentence that says why you are writing; a middle that holds only what the recipient needs to decide, with lists of 3 items at most; a closing with a clear action request and date; a signature from the details the user gave.
- Produce several tone versions: 3 by default (formal, friendly, short and direct); add a "cautious" version when the recipient is a first contact or more senior. Versions differ only in wording and framing; facts and the request stay identical.
- When the user wants a follow-up sequence, write 3 to 4 e-mails on the pattern "initial, gentle reminder on day 3, added value on day 7, final check-in on day 14", each with a new angle (extra material, an alternative, asking whether someone else is the right contact) and none repeating the previous wording.
- Deliver the result as a Markdown document with the tools available here (or directly in the reply if the user prefers), titled " E-mail Drafts".
- After delivering, recommend one version in a single sentence and remind the user to check the salutation, dates and attachments before sending.
Document structure
- One section per version: a level-2 heading with the version name, then subject / body / note (when to use this version).
- The follow-up sequence is its own section; each e-mail carries its suggested send time and trigger (no reply / replied but no action).
- Bilingual (the user's language and English) versions are separate blocks, not sentence-by-sentence pairs; the English is rewritten the way a native speaker would, not translated literally.
- Information the user must fill in uses bracketed placeholders such as [attachment name] or [date]; no blanks.
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.
- 4d ago First seen · 46 lines · 34 tokens per session scan A bb6fc82fc123
email-drafter is a skill published in the GitHub repository genspark-ai/genoffice (10,833 stars, last pushed 2d ago), licensed Apache-2.0. It adds 34 tokens to every session and 864 once invoked, about $0.0001 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.
bento-slides
Create and edit Bento presentations — single-file .bento.html decks whose document is plain JSON in a "#bento-doc" script block. Use whenever the user wants a slide deck or presentation: starting from NOTHING (it downloads the latest Bento app from bento.page automatically), from source material, or by improving an…
office-artifacts
Manage Office files, formats and versions; use Writer, Calc or Impress for content.
wafflebase
Use when the user asks about their Wafflebase documents — finding, reading, summarizing, creating, editing, moving, copying, commenting on or deleting a Wafflebase sheet, doc, slide deck, note, board or file. Covers setup, finding documents, the read and write protocol, links, and safety. Load the matching…
wafflebase-docs
Use for Wafflebase word-processor docs and markdown notes — reading or summarizing a doc, turning repository files into a doc or note, editing doc or note text, DOCX import/export, PDF export. Load the core wafflebase skill first for setup, links and the write protocol.
wafflebase-sheets
Use for Wafflebase spreadsheet work — reading ranges, analyzing or summarizing sheet data, writing cells and formulas, styles, rows/columns, tabs, charts, filters, pivots, CSV/JSON import and export, and recurring data loads into a sheet. Load the core wafflebase skill first for setup, links and the write protocol.