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 mathaix/OpenClawMachines --skill ocm-browsergit clone --depth 1 https://github.com/mathaix/OpenClawMachinesWrote 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/mathaix/openclawmachines/ocm-browser)<a href="https://agentmods.dev/skills/mathaix/openclawmachines/ocm-browser"><img src="https://agentmods.dev/badge/skills/mathaix/openclawmachines/ocm-browser/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/mathaix/openclawmachines/ocm-browser"><img src="https://agentmods.dev/badge/skills/mathaix/openclawmachines/ocm-browser.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 | $0.00051 | $0.00844 |
| Opus 5.5 | $0.00020 | $0.00338 |
| Sonnet 5.5 | $0.00010 | $0.00169 |
| Haiku 4.5 | $0.00005 | $0.00084 |
Grade A, and why
ocm-browser scanned grade A with 1 finding 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 1mo 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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
go through `bh` / `browser-harness`, not through raw `curl` / `websockets`. How it starts
The opening of the file, as written. The whole thing — 81 lines — stays where its author put it; the contents beside it link to each section on GitHub.
ocm-browser
OpenClaw machines don't run Chrome themselves. When you need a browser, a
separate browser VM is paired with your machine and its Chrome is reached
via a managed CDP bridge at http://192.168.100.1:9222. The browser-harness
skill gives you the Python helpers (click, js, screenshot, new_tab,
page_info, cdp) that talk to that Chrome.
Read this skill before writing browser code. Nothing below the readiness check below is usable unless the check passes.
Readiness check — ALWAYS run first
bh --check
Exit 0 means a browser VM is paired and reachable. Exit 1 means there is no browser to drive — stop, tell the user, do not claim you ran code you did not run.
The probe mirrors the endpoint bh will actually use, so a 0 exit here is a
reliable green light.
Minimum happy-path example
bh <<'PY'
new_tab("https://example.com")
wait_for_load()
info = page_info()
print(info["url"], info["title"])
PY
All helpers come from the browser-harness skill. bh wires up the bridge
CDP URL for you — you don't need to set BU_CDP_WS yourself.
When to consult interaction-skills
interaction-skills/pairing-readiness.md— ifbh --checkfails, or if the user asks "can I use a browser right now?"interaction-skills/bridge-proxy.md— if you need to understand what the bridge does, what network it exposes, SSRF policy, or why a request is blocked.interaction-skills/browser-vm-lifecycle.md— if the user asks about pairing/unpairing, what survives a VM restart, or how to restart Chrome without losing session state.interaction-skills/profile-persistence.md— if you need to save a logged-in browser session (LinkedIn, X/Twitter, Instagram, Facebook, most SaaS) and restore it after the browser VM gets recreated or a new one is paired.browser-harness/interaction-skills/cookies.md— generic Browser Harness cookie guidance.
Important constraints
- Cookies and session state live in the browser VM, not this machine.
They persist while the browser VM runs; they vanish if the user unpairs
or recreates it. To keep them across unpair/recreate, save the cookie
jar first — see
interaction-skills/profile-persistence.md. - The bridge has a fail-closed SSRF policy. Arbitrary internal IPs are blocked; the bridge's own RFC1918 address is allowlisted by the platform.
- You do not have direct TCP access to the browser VM's Chrome. Always
go through
bh/browser-harness, not through rawcurl/websockets.
What ships with it
4 files 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.
- 1mo ago First seen · 81 lines · 51 tokens per session scan A 74daeb977700
ocm-browser is a skill published in the GitHub repository mathaix/OpenClawMachines (60 stars, last pushed 3mo ago), licensed Apache-2.0. It adds 51 tokens to every session and 844 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
browser-use
Direct browser control via CDP for web interaction: automation, scraping, testing, screenshots, and site/app work. Also covers hosted cloud browsers for AI agents: the browser-use Python library (Browser(usecloud=True)), Playwright, Puppeteer or any CDP client.
x402
Set up Browser Use Cloud payments with x402 — pay per request from a crypto wallet (USDC on Base mainnet), no signup or API key. Two setups it works out up front — "just use it" (set up a wallet so you or Claude Code can run cloud browser tasks paid from the wallet — Claude writes and runs throwaway scripts, nothing…
blocked-page-recovery
Use when a fetch fails: 403/429, paywall, WAF, bot wall.
playwright
Use when the task requires capturing or automating a real browser from the terminal.
playwright-cli
Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.
web-search
This skill should be used when users need to search the web for information, find current content, look up news articles, search for images, or find videos. It uses DuckDuckGo's search API to return results in clean, formatted output (text, markdown, or JSON). Use for research, fact-checking, finding recent…