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 agentmods add skills/markus-global/markus/chrome-devtoolsnpx skills add markus-global/markus --skill chrome-devtoolsgit clone --depth 1 https://github.com/markus-global/markusWhat 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 | $0.00017 | $0.04098 |
| Opus 5 | $0.00009 | $0.02049 |
| Sonnet 5 | $0.00003 | $0.00820 |
| Haiku 4.5 | $0.00002 | $0.00410 |
Grade B, and why
chrome-devtools scanned grade B with 2 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 2d 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.
Asks for rootmediumPrivilege escalation
A mod that escalates privileges can change anything on the machine, not only the project.
- **Linux (Debian/Ubuntu)**: `wget -q -O - https://dl.google.com/linux/linux_signing_key.pub | sudo apt-key add - && sudo apt update && sudo apt install google-chrome-stable` Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
- **Linux (Debian/Ubuntu)**: `wget -q -O - https://dl.google.com/linux/linux_signing_key.pub | sudo apt-key add - && sudo apt update && sudo apt install google-chrome-stable` How it starts
The opening of the file, as written. The whole thing — 373 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Chrome DevTools Browser Automation
You have access to a live Chrome browser via Chrome DevTools MCP tools (prefixed chrome-devtools__).
This gives you full browser automation: navigation, clicking, typing, screenshots, JS evaluation,
network inspection, and performance profiling.
When to use these tools
Use Chrome DevTools tools when you need to:
- Interact with web pages (click buttons, fill forms, navigate)
- Test web applications in a real browser
- Take screenshots of pages or elements
- Inspect console logs, network requests, or page structure
- Run Lighthouse audits or performance traces
- Debug frontend issues with live DOM inspection
Do NOT use these tools when web_fetch or web_search suffice (simple content retrieval or search).
Chrome DevTools is for interactive browser sessions that require a real rendering engine.
Connection Modes
Markus supports three ways to connect to Chrome, listed from best to fallback:
Mode 1: Markus Chrome Extension (recommended)
The Markus Browser Automation Chrome extension provides the smoothest experience:
- No debugging dialog — the extension uses
chrome.debuggerAPI internally - Works when screen is locked or sleeping — no OS-level interaction needed
- Cross-platform — works on macOS, Windows, and Linux identically
- Instant startup — no
npxdownload, no child process spawn - No extra permissions — no macOS Accessibility or Windows UI Automation needed
How to install:
- Build the extension (one-time):
cd packages/chrome-extension && pnpm install && pnpm run build - Open Chrome →
chrome://extensions/ - Enable Developer mode (toggle in top-right)
- Click Load unpacked → select
packages/chrome-extension/dist - The Markus icon appears in the toolbar
How it works:
When Markus starts, it launches a WebSocket bridge on ws://127.0.0.1:9333. The extension
auto-connects to this bridge. All browser tool calls are routed through the extension instead
of spawning an external MCP process.
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.
- 2d ago First seen · 373 lines · 17 tokens per session scan B c77906ee46b8
chrome-devtools is a skill published in the GitHub repository markus-global/markus (157 stars, last pushed 5d ago), licensed Apache-2.0. It adds 17 tokens to every session and 4,098 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it B with 2 findings (asks for root, 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
openbot-data-access
Governs how the OpenBot browser app reads and writes server data — every request goes through client in app/src/lib/client.ts, every read is a queryOptions factory in app/src/lib/ /queries.ts, every write is a mutationOptions factory in app/src/lib/ /mutations.ts, and components consume them through…
openbot-screen-layout
The default layout for every OpenBot configuration screen — PageShell and its prose/wide widths, PageSection and PageRows, Item row composition, the settings-row pattern where a summary and a chevron open a dialog, and the size and variant vocabulary. This is what a new screen looks like unless an instruction says…
oauth-setup
Connect an OAuth 2.0 service (Gmail, Google APIs, etc.) using the built-in oauthauthorize tool — get a click-to-authorize link, auto-capture the code, and seal the tokens. Use whenever the operator asks to connect Gmail / a Google API / any OAuth account.
tmux
Remote-control tmux sessions for interactive CLIs by sending keystrokes and scraping pane output.
seo-crawl
Agent-wide core connector rule: when any connected search console is used, discover its operations with listconnectortools and invoke each selected operation through callconnectortool; naming listsites or an analytics operation without the core invocation is not execution.
seo-opportunity
For product-focused strategy, prefer observed use-case, pricing, comparison, and trust/docs pages. Name missing page types as coverage gaps and map recommendations to owned pages, first-party proof, answer blocks, and one cannibalization owner per overlapping query cluster.