Getting it into your agent
It runs from inside its repository, so the clone comes first — what it calls does not travel with the file alone.
git clone --depth 1 https://github.com/ironbee-ai/ironbee-devtools-skillsnpx agentmods add skills/ironbee-ai/ironbee-devtools-skills/ironbee-browser-devtools-cliWrote 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/ironbee-ai/ironbee-devtools-skills/ironbee-browser-devtools-cli)<a href="https://agentmods.dev/skills/ironbee-ai/ironbee-devtools-skills/ironbee-browser-devtools-cli"><img src="https://agentmods.dev/badge/skills/ironbee-ai/ironbee-devtools-skills/ironbee-browser-devtools-cli.svg" alt="Measured on agentmods" 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 | $0.00150 | $0.03905 |
| Opus 5 | $0.00075 | $0.01953 |
| Sonnet 5 | $0.00030 | $0.00781 |
| Haiku 4.5 | $0.00015 | $0.00391 |
Grade A, and why
ironbee-browser-devtools-cli 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 6d 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 — 348 lines — stays where its author put it; the contents beside it link to each section on GitHub.
IronBee Browser DevTools CLI
Command-line interface for browser automation, debugging, and testing using Playwright. Part of IronBee DevTools. This CLI targets the browser platform; for Node.js backend debugging see ironbee-node-devtools-cli, and for HTTP/gRPC/GraphQL/WS + log + database verification see ironbee-backend-devtools-cli.
ironbee-browser-devtools-cli and ironbee-devtools-cli are aliases — both ship with the same package and target the browser platform by default. Examples in this doc use ironbee-browser-devtools-cli; substitute ironbee-devtools-cli if you prefer the shorter name.
Installation
# Install this skill (skills.sh)
npx skills add ironbee-ai/ironbee-devtools-skills
# Install the CLI binary
npm install -g @ironbee-ai/devtools
Requires Node.js >= 22.16.0 (engines.node in the package): the backend db domain's SQLite engine and the FTS5 scenario-search strategy run on Node's built-in node:sqlite, which replaced the last native dependency.
Quick Start
# Navigate to a URL
ironbee-browser-devtools-cli navigation go-to --url "https://example.com"
# Take a screenshot
ironbee-browser-devtools-cli content take-screenshot --name "homepage"
# Get page content as text
ironbee-browser-devtools-cli content get-as-text
Ref-based workflow (recommended for AI agents): Call a11y take-aria-snapshot first to get refs (e1, e2, ...), then use refs in interaction click --selector "e1" or content take-screenshot --annotate for numbered element labels.
Global Options
| Option | Description | Default |
|---|---|---|
--port <number> |
Daemon server port | 2020 |
--session-id <string> |
Session ID for browser state persistence | auto |
--json |
Output results as JSON (recommended for AI) | false |
--quiet |
Suppress log messages | false |
--verbose |
Enable debug output | false |
--timeout <ms> |
Operation timeout | 30000 |
--headless |
Run browser in headless mode | true |
--no-headless |
Run browser with visible window | - |
--persistent |
Preserve cookies/localStorage | false |
--no-persistent |
Clear state on session end | - |
--user-data-dir <path> |
Browser user data directory | OS temp |
--use-system-browser |
Use system Chrome | false |
--browser-path <path> |
Custom browser path | auto |
What ships with it
14 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.
- references/a11y.md 5.5 KB
- references/content.md 8.3 KB
- references/debug.md 7.6 KB
- references/execute.md 5.4 KB
- references/figma.md 7.0 KB
- references/interaction.md 9.9 KB
- references/jira.md 11 KB
- references/linear.md 12 KB
- references/navigation.md 5.5 KB
- references/o11y.md 15 KB
- references/react.md 6.9 KB
- references/scenario.md 31 KB
- references/stub.md 6.8 KB
- references/sync.md 4.2 KB
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.
- 6d ago First seen · 348 lines · 0 tokens per session scan A 6eda53832c65
ironbee-browser-devtools-cli is a skill published in the GitHub repository ironbee-ai/ironbee-devtools-skills (3 stars, last pushed 1mo ago), licensed MIT. It adds 150 tokens to every session and 3,905 once invoked, about $0.0007 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 2026-08-31.
Other skills, from other repositories
playwright-tabbed-orchestration
Parent Agent uses playwright-tabbed to allocate browser tabs and partition tasks, then spawns child agents in parallel — each operating on its assigned tabid — for browser acceptance testing or automation, with a final aggregated summary. Suitable for multi-page validation, parallel module checks, and…
e2e-testing-patterns
Master end-to-end testing with Playwright and Cypress to build reliable test suites that catch bugs, improve confidence, and enable fast deployment. Use when implementing E2E tests, debugging flaky tests, or establishing testing standards.
browser-testing-with-devtools
Tests in real browsers via Chrome DevTools MCP. Use when building or debugging anything that runs in a browser. Use when you need to inspect the DOM, capture console errors, analyze network requests, profile performance, or verify visual output with real runtime data. Requires the chrome-devtools MCP server to be…
actionbook-web-test
Run browser-based web tests against websites using Actionbook CLI. Activate when the user wants to test a website workflow, run smoke tests, verify a user flow, check if a web application works, run regression tests, or validate browser-based interactions. Supports test definition, execution, assertion, reporting, and…
browser-viz-verify
Verifies that a NetClaw-generated visualization HTML file (three.js, canvas, drawio, UML, markmap) actually renders correctly — screenshot, console-error check, and an optional Lighthouse audit. Use immediately after generating any browser-based visualization output, to close the QA gap that otherwise requires a human…
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.