Problem Category: Browser automation for creating and editing Tencent Docs (腾讯文档) online spreadsheets.
Applicable Questions: Creating new spreadsheets on Tencent Docs, filling in cell data (headers and data rows), getting document share links.
Key Features:
Uses Playwright with persistent browser context for session management
Phased approach: navigate → click 新建 → click 表格 → click 空白表格 in iframe template picker → fill data → verify → get share link
Handles the template picker iframe (tmall-panel-iframe) which appears after clicking 表格
Uses name box (input.bar-label) for reliable cell navigation
Uses formula bar (div.formula-input, contenteditable) for reliable data input into canvas-based cells
Verifies cell content by reading formula bar after navigation
Handles popup windows/new tabs for newly created documents
Skills Used: playwright, google_search, execute_shell_command
Reasoning Pattern: LLM parses headers/data from query → navigates to docs.qq.com → clicks New → selects Spreadsheet → clicks Blank Spreadsheet in iframe → navigates to each cell via name box → types via formula bar → verifies each cell → returns share link
Input Format: Natural language query describing what spreadsheet to create and what data to fill in.
Output Format: Dict with 'answer' containing share link and completion message, 'summary' with detailed trace.
Critical Implementation Details:
Tencent Docs spreadsheet is canvas-based - cannot type directly into cells via keyboard
Must click formula bar (div.formula-input) to focus it before typing
After clicking 表格 in creation menu, a template picker appears in iframe (tmall-panel-iframe) - must click 空白表格 inside the iframe
Cell navigation: click name box (input.bar-label) → triple-click to select → type cell ref (e.g. A1) → Enter
Data input: navigate to cell → click formula bar → type text → Enter to commit
Skills Used
playwright
Usage
Entry file: tencent_docs_final.py
Query type: Pass a focused sub-question as the query.
1---2name: zzatpku-agentfactory-agentfactory3description: tencent_docs_final4---56# tencent_docs_final78## Description9**Problem Category**: Browser automation for creating and editing Tencent Docs (腾讯文档) online spreadsheets.10**Applicable Questions**: Creating new spreadsheets on Tencent Docs, filling in cell data (headers and data rows), getting document share links.11**Key Features**:12- Uses Playwright with persistent browser context for session management13- Phased approach: navigate → click 新建 → click 表格 → click 空白表格 in iframe template picker → fill data → verify → get share link14- Handles the template picker iframe (tmall-panel-iframe) which appears after clicking 表格15- Uses name box (input.bar-label) for reliable cell navigation16- Uses formula bar (div.formula-input, contenteditable) for reliable data input into canvas-based cells17- Verifies cell content by reading formula bar after navigation18- Handles popup windows/new tabs for newly created documents19**Skills Used**: playwright, google_search, execute_shell_command20**Reasoning Pattern**: LLM parses headers/data from query → navigates to docs.qq.com → clicks New → selects Spreadsheet → clicks Blank Spreadsheet in iframe → navigates to each cell via name box → types via formula bar → verifies each cell → returns share link21**Input Format**: Natural language query describing what spreadsheet to create and what data to fill in.22**Output Format**: Dict with 'answer' containing share link and completion message, 'summary' with detailed trace.23**Critical Implementation Details**:241. Tencent Docs spreadsheet is canvas-based - cannot type directly into cells via keyboard252. Must click formula bar (div.formula-input) to focus it before typing263. After clicking 表格 in creation menu, a template picker appears in iframe (tmall-panel-iframe) - must click 空白表格 inside the iframe274. Cell navigation: click name box (input.bar-label) → triple-click to select → type cell ref (e.g. A1) → Enter285. Data input: navigate to cell → click formula bar → type text → Enter to commit2930## Skills Used31playwright3233## Usage3435**Entry file**: `tencent_docs_final.py`3637**Query type**: Pass a focused sub-question as the query.3839**How to call**:40```xml41<action>run_subagent</action>42<params>{"skill_name": "tencent_docs_final", "query": "<your focused sub-question>"}</params>43```4445---46> Source: [zzatpku/AgentFactory](https://github.com/zzatpku/AgentFactory) — distributed by [TomeVault](https://tomevault.io).47<!-- tomevault:4.0:skill_md:2026-07-03 -->
Run npx skillmds@latest add tomevault-io/zzatpku-agentfactory-agentfactory in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.
tencent_docs_final It is listed under Coding & Dev Tools on SkillMD.
This skill has not completed SkillMD's automated safety review yet. Independent scanners report: SkillSpector: PASS, Skill Scanner: PASS. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.
This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.
Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.
tomevault-io (@tomevault-io) published this skill. Their other Agent Skills are listed on their SkillMD profile.
tomevault-io