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 zj-unicom-ai/UniEmployee --skill tencent-ima-wikigit clone --depth 1 https://github.com/zj-unicom-ai/UniEmployeeWrote 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/zj-unicom-ai/uniemployee/tencent-ima-wiki)<a href="https://agentmods.dev/skills/zj-unicom-ai/uniemployee/tencent-ima-wiki"><img src="https://agentmods.dev/badge/skills/zj-unicom-ai/uniemployee/tencent-ima-wiki/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/zj-unicom-ai/uniemployee/tencent-ima-wiki"><img src="https://agentmods.dev/badge/skills/zj-unicom-ai/uniemployee/tencent-ima-wiki.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 | $0.00127 | $0.01474 |
| Opus 5.5 | $0.00051 | $0.00590 |
| Sonnet 5.5 | $0.00025 | $0.00295 |
| Haiku 4.5 | $0.00013 | $0.00147 |
Grade A, and why
tencent-ima-wiki scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 17d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 169 lines — stays where its author put it; the contents beside it link to each section on GitHub.
腾讯 ima 知识库内容获取
概述
通过浏览器自动化连接已登录的 ima.qq.com/wikis 页面,从 React 组件内部状态中提取文章的 sourcePath(原始微信公众号等外部链接),再逐个打开新标签页获取文章正文。
前置条件
- 用户已在 Chrome 中打开 https://ima.qq.com/wikis 并完成登录
- 当前选中了目标知识库(如"万仁刚的知识库")
- 使用
mac_computer_use_tool配合plane="bu"进行浏览器操作
工作流程
第一步:连接浏览器并确认页面状态
import seed_browser_use as bu
# 同步连接浏览器
bu.resync()
# 确认页面已加载
bu.wait_for_load(timeout=15)
page_info = bu.page_info()
print("当前页面:", page_info.get("url"))
如果 resync 失败或没有活动页面,直接用 bu.navigate("https://ima.qq.com/wikis") 打开页面。
第二步:提取文章列表的 sourcePath
ima 知识库使用虚拟滚动列表,文章数据存储在 React Fiber 内部状态中,DOM 元素上没有直接的链接属性。需要通过遍历 React Fiber 提取。
使用 scripts/extract_source_paths.js 中的代码逻辑,提取前 N 条文章的标题和原始链接:
js_code = """
(async () => {
const results = [];
const count = arguments[0] || 10; // 要提取的条数
for (let i = 0; i < count; i++) {
const item = document.querySelector(`[data-index="${i}"]`);
if (!item) continue;
const fiberKey = Object.keys(item).find(k => k.startsWith('__reactFiber'));
let fiber = item[fiberKey];
let foundPath = null;
let foundTitle = null;
let depth = 0;
function searchObj(obj, path, depth2) {
if (!obj || depth2 > 8 || foundPath) return;
if (typeof obj !== 'object') return;
for (let key in obj) {
try {
if ((key === 'sourcePath' || key === 'source_path') && typeof obj[key] === 'string') {
foundPath = obj[key];
}
if ((key === 'title' || key === 'name') && typeof obj[key] === 'string' && obj[key].length > 5) {
foundTitle = obj[key];
}
if (typeof obj[key] === 'object' && obj[key] !== null) {
searchObj(obj[key], path + '.' + key, depth2 + 1);
}
} catch(e) {}
}
}
while (fiber && depth < 30 && !foundPath) {
if (fiber.memoizedProps) {
searchObj(fiber.memoizedProps, 'props', 0);
}
fiber = fiber.return;
depth++;
}
results.push({
index: i,
title: foundTitle || item.innerText.split('\\n')[0],
sourcePath: foundPath
});
}
return results;
})()
"""
# 传入要提取的条数
result = bu.js(js_code.replace('arguments[0] || 10', '5'))
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 17d ago First seen · 169 lines · 127 tokens per session scan A 99ed9b144748
tencent-ima-wiki is a skill published in the GitHub repository zj-unicom-ai/UniEmployee (362 stars, last pushed 8d ago), licensed MIT. It adds 127 tokens to every session and 1,474 once invoked, about $0.0005 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 2026-09-29.
Other skills, from other repositories
webshop-result-page-navigator
Navigates through paginated search results by moving to the next or previous page. Use this skill when the current page of product listings does not contain a suitable match and you need to review more options. It identifies and clicks the appropriate pagination controls (e.g., 'next >' or '< prev') to load additional…
webshop-search-executor
Executes a search on an e-commerce platform using parsed keywords. Use when you have extracted search terms from a user query and need to perform the actual search[] action on the WebShop interface. Takes structured search terms and performs a search action, returning a list of product results for evaluation.
firecrawl
Scrape pages, crawl public sites, map URLs, and run JSON-schema extraction through managed or self-hosted Firecrawl.
lingshu-browser
A routing guide for browser and web-development topics, including rendering, the DOM, event loops, web storage, content security, performance, and progressive web apps. It directs a task to a more specific unit when needed.
browser-440d4398
A browser popup controller that allows popups started by a user action and blocks and records automatic popups.
browser-02898b15
A narrowly targeted browser skill for event bubbling, where a browser event travels from the element that received it up through its parent elements. It is intended only when the task explicitly concerns this event path.