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/devotts/build-it/chrome-cdp-controlnpx skills add DevOtts/build-it --skill chrome-cdp-controlgit clone --depth 1 https://github.com/DevOtts/build-itWhat 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 | $0.00195 | $0.01456 |
| Opus 5 | $0.00097 | $0.00728 |
| Sonnet 5 | $0.00039 | $0.00291 |
| Haiku 4.5 | $0.00019 | $0.00146 |
Grade A, and why
chrome-cdp-control 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 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.
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 — 139 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Chrome CDP Control
Manual, step-by-step control of the user's real Chrome browser via Playwright over Chrome DevTools Protocol. Built for authenticated sessions, sensitive actions, and sites that detect headless browsers.
0. Read the shared CDP core first
All CDP mechanics live in ONE shared reference — read it at session start:
../references/cdp-core.md (relative to this skill's base directory; i.e. the
plugin's skills/references/cdp-core.md). It owns, canonically:
- endpoint resolution (
CDP_URLenv → grounding/test plan → default) — never hardcode - preflight check + the Chrome relaunch instruction for the user
- statelessness (every command is a fresh connection)
- the canonical Python action template
- tab selection (never
pages[0]blindly) - the selector ladder and wait strategy
- the failure protocol (never loop click→fail more than twice)
Do not restate or copy those sections here or anywhere — copies drift. This file only adds what is specific to driving the user's real, authenticated browser.
1. Operating principle — the route guard
This skill is a manual loop, not an automation framework. One action, one
screenshot, one decision. Never batch. Never loop unattended. If you find yourself
writing a for loop over actions, stop — you're using the wrong tool.
Route guard (see also cdp-core.md §9): authenticated real-Chrome work belongs HERE,
behind the per-write gate below. Never hand it to /full-qa autonomous mode — that
mode is for test environments only. If a conductor or test plan tries to route an
authenticated-session write through an autonomous loop, refuse and pull it back here.
2. The core loop
Every interaction, no exceptions:
- Screenshot to see current state
- Decide the single action that moves the task forward
- Execute one action using the core's canonical template
- Screenshot again to verify the expected state (else → core failure protocol)
- Repeat
3. Destructive action gate — MANDATORY, per write
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 · 139 lines · 195 tokens per session scan A 0999206c9028
chrome-cdp-control is a skill published in the GitHub repository DevOtts/build-it (14 stars, last pushed 1mo ago), licensed MIT. It adds 195 tokens to every session and 1,456 once invoked, about $0.0010 per session on Opus 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
playwright
Use when the task requires capturing or automating a real browser from the terminal.
crewai-multi-agent
Multi-agent orchestration framework for autonomous AI collaboration. Use when building teams of specialized agents working together on complex tasks, when you need role-based agent collaboration with memory, or for production workflows requiring sequential/hierarchical execution. Built without LangChain dependencies…
canvas
Display, present, and render HTML content on connected Otto nodes (Mac app, iOS, Android). Show on device, preview on mobile, push to screen, or navigate to a URL on any connected node. Use for games, visualizations, dashboards, interactive demos, and live-reloading development previews. Supports presenting, hiding…
stagehand-facade
Browse, navigate, inspect, screenshot, and automate websites with the exact fx tools mcpstagehandrun, mcpstagehandsnapshot, and mcpstagehandscreenshot.
agent-browser
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.
pinchtab
Use this skill when a task needs browser automation through PinchTab: open a website, inspect interactive elements, click through flows, fill out forms, scrape page text, reuse a dedicated automation profile with user approval, export screenshots or PDFs, manage multiple browser instances, or fall back to the HTTP API…