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 rexleimo/aios --skill skill-constraintsgit clone --depth 1 https://github.com/rexleimo/aiosWrote 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/rexleimo/aios/skill-constraints)<a href="https://agentmods.dev/skills/rexleimo/aios/skill-constraints"><img src="https://agentmods.dev/badge/skills/rexleimo/aios/skill-constraints/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/rexleimo/aios/skill-constraints"><img src="https://agentmods.dev/badge/skills/rexleimo/aios/skill-constraints.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00048 | $0.01434 |
| Opus 5.5 | $0.00019 | $0.00574 |
| Sonnet 5.5 | $0.00010 | $0.00287 |
| Haiku 4.5 | $0.00005 | $0.00143 |
Grade A, and why
skill-constraints 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 29d 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 — 135 lines — stays where its author put it; the contents beside it link to each section on GitHub.
浏览器操作约束
Working directory: any (MCP tools are session-relative; file paths are relative to project root)
所有浏览器自动化操作 MUST 遵循此约束规则，确保安全、高效、可追溯。
When to Use
- 执行任何浏览器自动化操作时
- 使用 MCP 工具（
page.extract_text,page.get_html,page.screenshot等）时 - 进行任何运营操作（发布笔记、点赞、评论等）时
Core Pattern
浏览器操作
0. 启动浏览器时默认要求可见界面并走 CDP
- 优先：`chrome.launch_cdp { port: 9222, user_data_dir: '~/.chrome-cdp-profile' }`
- 随后：`browser.connect_cdp { cdp_url: 'http://127.0.0.1:9222' }`
- 仅在无图形环境或后台 smoke test 时才允许 headless 模式
- 当同一 `userDataDir` 被其他进程占用时，优先复用已启动浏览器，不共享被锁目录
- 若目标是“多个 agent 共享同一登录态”，统一连接同一个 CDP endpoint
1. 优先使用文本/DOM证据做决策
- 先读 `page.extract_text`，必要时补充 `page.get_html`
- 比整页截图更高效，可快速定位按钮文案和页面状态
- 不使用 `chrome-devtools` 工具链执行业务流程，统一走 `mcp-browser-use` 的 browser-use 工具链（`chrome.*` / `browser.*` / `page.*`）
2. 只有视觉信息不足时才截图，并保存到 temp/ 目录
- 条件：文本/HTML 证据不足以判断状态
- 使用：`page.screenshot`
- 路径：`aios/temp/{操作类型}_{时间戳}.png`
- 示例：`login_20240301_120000.png`
- 若 `aios/temp/` 不存在则先创建：`mkdir -p aios/temp`
操作间隔
Every browser action (click, type, navigate) MUST be followed by a random wait. No exceptions.
Use page.wait with a random duration, or if running in a shell context:
sleep $((RANDOM % 26 + 5))
In MCP context, use page.wait({ duration: <random 5-30s> }) or equivalent.
This is not optional — skipping intervals risks detection and account bans. If you batch operations without waiting, the entire task is considered failed regardless of outcome.
Rules
MUST NOT（禁止行为）
| 行为 | 原因 |
|---|---|
| 直接在对话中粘贴大段截图 | 浪费 token，MUST 保存到文件 |
| 跳过反检测 | 每次操作前 MUST 执行 |
| 跳过操作间隔 | MUST 随机 5-30 秒 |
| 在非 temp 目录保存截图 | MUST 保存到 aios/temp/ |
| 自动化执行第三方平台登录 | 登录 MUST 由人工完成（含 2FA） |
MUST（必需行为）
| 行为 | 原因 |
|---|---|
| 操作前执行反检测 | 使用反检测脚本（如项目中已有）或通过 CDP 指纹配置规则 |
| 截图保存到 temp/ | 路径固定为 aios/temp/ |
| 先读文本/DOM | page.extract_text → page.get_html |
| 使用 grep 搜索快照 | 先保存 snapshot 到文件，再用 grep 搜索 |
| 记录到历史 | 关键操作写入 .aios/context-db checkpoint/event |
| 登录态检测 | 识别到登录页/验证码/2FA 时 MUST 先提示用户协作登录 |
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.
- 29d ago Changed 1a261820c0b9
- 1mo ago Changed 7a51d86b1160
- 1mo ago Changed beb55e80ebc0
- 1mo ago Changed 8fc56ce2b6aa
- 1mo ago First seen · 135 lines · 48 tokens per session scan A 4aea2515b980
skill-constraints is a skill published in the GitHub repository rexleimo/aios (54 stars, last pushed yesterday), licensed MIT. It adds 48 tokens to every session and 1,434 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 2026-08-30.
Other skills, from other repositories
opencli-autofix
Automatically fix broken OpenCLI adapters when commands fail. Load this skill when an opencli command fails — it guides you through collecting a trace artifact, patching the adapter, retrying, and filing an upstream GitHub issue after a verified fix. Works with any AI agent.
opencli-adapter-author
Build or improve an OpenCLI JS adapter using observed site behavior, existing adapters and browser API evidence. For ad-hoc browser interaction, see opencli-browser.
opencli-browser
Drive Chrome with OpenCLI to inspect pages, interact with logged-in sites, capture API evidence, and manage native tabs, groups and windows. For reusable adapters, see opencli-adapter-author.
webcmd-browser
Use when a live browser task requires Playwright interaction, authenticated handoff, visible UI verification, or ad-hoc page inspection.
surf
Control Chrome browser via CLI for testing, automation, and debugging. Use when the user needs browser automation, screenshots, form filling, page inspection, network/CPU emulation, DevTools streaming, or AI queries via ChatGPT/Gemini/Perplexity/Grok/AI Studio.
web-browsing-cli
Token-efficient web browsing and web content extraction for AI agents. Use when reading a URL, browsing websites, checking links, extracting static page content, or replacing raw HTML and browser screenshots.