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 ClawCap/ManoBrowser --skill browser-automationgit clone --depth 1 https://github.com/ClawCap/ManoBrowserWrote 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/clawcap/manobrowser/browser-automation)<a href="https://agentmods.dev/skills/clawcap/manobrowser/browser-automation"><img src="https://agentmods.dev/badge/skills/clawcap/manobrowser/browser-automation/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/clawcap/manobrowser/browser-automation"><img src="https://agentmods.dev/badge/skills/clawcap/manobrowser/browser-automation.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 | $0.00103 | $0.07324 |
| Opus 5.5 | $0.00041 | $0.02930 |
| Sonnet 5.5 | $0.00021 | $0.01465 |
| Haiku 4.5 | $0.00010 | $0.00732 |
Grade A, and why
browser-automation 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 2mo 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 — 875 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Browser Automation Skill
Complete browser automation workflow based on MCP Browser Server, providing stable and reliable web page operations.
Core Capabilities
- CDP-Powered Automation: Chrome DevTools Protocol for reliable, background operations
- Smart Navigation System: Auto-detect open tabs, avoid redundant navigation
- Advanced Element Interaction: UID-based targeting, accessibility tree analysis
- Content Extraction: HTML/text extraction, saved locally
- Network Requests: Send HTTP requests with browser cookies
- Advanced Features: File upload, script injection, history navigation
📖 Quick Start
🎯 Execution Flow
Standard Workflow (CDP-Powered, Recommended)
1. get_windows_and_tabs (Check browser state)
2. chrome_navigate(url) (Navigate to target page, background mode)
3. chrome_screenshot(tabId) (Verify loading)
4. [Optional] chrome_page_protection_enable (Protect from user interference)
5. [Optional] chrome_accessibility_snapshot (Get interactive elements with UIDs)
6. [Optional] chrome_click_element(uid) (Click using UID - most reliable)
7. chrome_get_web_content(tabId) (Extract content)
8. [Optional] chrome_page_protection_disable (Restore user control)
9. chrome_close_tabs (Clean up: close opened tabs after completion)
⚠️ User Intervention: When encountering login/CAPTCHA/complex decisions, use mcp__user_interaction_tools__ask_user_confirmation:
Example parameters:
{
"type": "user_takeover",
"message": "Detected login required. Please complete login and click confirm.",
"confirm_text": "Continue",
"cancel_text": "Cancel",
"tabId": "1625463389",
"timeout": 300
}
CDP Advantages:
- Background execution: No tab activation, faster performance
- UID-based targeting: Most reliable element location method
- Better error handling: CDP provides detailed error information
- Persistent state: Maintains page state across operations
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.
- 2mo ago First seen · 875 lines · 103 tokens per session scan A 30ff8f40e151
browser-automation is a skill published in the GitHub repository ClawCap/ManoBrowser (9 stars, last pushed 6mo ago), licensed MIT. It adds 103 tokens to every session and 7,324 once invoked, about $0.0004 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 2026-08-31.
browser-automation compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| browser-automationClawCap/ManoBrowser | Skill | 9 | 6mo ago | MIT | A | 103 | Claude Code, Codex |
| agent-browser-clisleepinginsummer/agent-browser-cli | Skill | 646 | 5d ago | MIT | A | 32 | Claude Code, Codex |
| aipex-browserAIPexStudio/AIPex | Skill | 1.3k | 2mo ago | MIT | A | 61 | Claude Code, Codex |
| browser-bridgekoltyakov/browser-bridge | Skill | 29 | yesterday | MIT | A | 49 | Codex |
| codex-browserDeliciousBuding/codex-browser-bridge | Skill | 25 | 5d ago | MIT | A | 41 | Claude Code |
| browser-relayreliefeai/browser-relay | Skill | 138 | 1mo ago | MIT | A | 52 | Claude Code |
| real-browser-controlofershap/real-browser-mcp | Skill | 58 | 4d ago | MIT | A | 158 | Claude Code, Codex |