Agent Skills · Office Efficiency
Curated skills for Claude, Cursor and other AI agents. Copy one prompt to install.
Office skills promise to create, edit, or summarize documents, spreadsheets, and slides through an agent. The honest version of that promise is a prompt that drives one document suite you already pay for. The risky version offers to "clean up the drive."
Start here when the artifact is a doc, a sheet, or a deck, and you can name the file. A Tencent Docs style integration belongs on this shelf because the job is the document API, not the prose. A writing skill that only drafts text you paste into a doc belongs in content creation instead. If the card's prompt never mentions a document type, it was filed here because the title said productivity.
Read the prompt for scope before you read the download count. A shared company drive plus an instruction to reorganize folders will touch files other people own. The first run belongs in a scratch document in a personal folder. If the skill wants an API key, use a key that can reach only that scratch file. When not to install a skill is the longer version of that rule, and the office shelf is where people skip it because the task feels clerical.
Sort this category by recently updated when you depend on a vendor API. Document tools break when the vendor renames an endpoint, and a high download count often means the listing is old and popular, not that last month's API still matches the prompt. How ranking works explains why downloads float old cards to the top.
Skip a card that claims to attend meetings, send mail, and file expenses in one prompt. That is three permissions wearing one name. Pick the document operation you actually need this week. Editor's notes on this shelf, when present, separate the document job from the key it requires. A card without a note is still only the publisher's description.
Do not use an office skill as a system of record. The document in the suite is the record. The skill is a way to draft or extract. If the extract disagrees with the sheet, the sheet wins until you check the cells yourself.
The complete set of online document operation tools provided by Tencent Docs MCP, supporting creation, querying, and editing of smart docs, Excel, PPT, mind maps, and more.
Tencent Cloud Table Recognition V3 is an OCR skill for detecting and recognizing tables in images or PDFs, supporting various table types like linear and borderless tables, with Excel export.
Generate and edit .pptx decks with python-pptx, applying structured layouts, design rules, native charts, and visual QA to reduce template-like output.
Create, repair, and recalculate xlsx, xlsm, csv, or tsv workbooks with formula-safe edits, validation, and recalculation.
Interprets tender files, generates bid docs, and checks compliance via the Bailian API.
Convert HTML files or URLs into PDFs that preserve CSS, support Chinese, and handle automatic pagination.
Turn post-meeting text into bilingual minutes, preserving structure and consistent terminology.
Convert teaching schedules from talent development plan PDFs into structured Excel files, with cross-page tables, merged cells, multi-semester course splitting, major metadata extraction, and batch output.
Generate professional .pptx presentations from a topic or outline, with themes, common layouts, speaker notes, and 16:9/4:3 support.
Generates a fixed-structure A-share tech daily across five tracks, screening 24-hour news and selecting leader and growth stocks.
Generate teaching PPT outlines, image prompts, and python-pptx scripts for higher-vocational interior design and construction management courses.
An OCR text recognition service for agents, supporting URL/Base64 images, async task polling, and per-call billing.