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 agents/serkan-ozal/browser-devtools-claude/defaultgit clone --depth 1 https://github.com/serkan-ozal/browser-devtools-claudeWhat 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 | $0.00026 | $0.01242 |
| Opus 5 | $0.00013 | $0.00621 |
| Sonnet 5 | $0.00005 | $0.00248 |
| Haiku 4.5 | $0.00003 | $0.00124 |
Grade A, and why
default 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 — 86 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Browser DevTools MCP — Default Agent
When using Browser DevTools MCP, follow these rules so that page structure, interactions, and batch execution are done correctly.
*** CRITICAL — TO UNDERSTAND PAGE STRUCTURE (do not default to screenshot) ***
Use ARIA snapshot first, then AX tree if needed. Use screenshot only when necessary.
After navigating to a page, your FIRST inspection call MUST be a11y_take-aria-snapshot (or a11y_take-ax-tree-snapshot), NOT content_take-screenshot.
DO NOT take a screenshot just to "see the page" or understand what is on it.
- Call
a11y_take-aria-snapshotFIRST for structure, roles, and semantics. - If you need layout, bounding boxes, or occlusion: call
a11y_take-ax-tree-snapshot. - Call
content_take-screenshotONLY when you need to verify how something looks visually (e.g. design check, visual bug, or when ARIA/AX are insufficient). Do not use screenshot to understand page structure.
*** CRITICAL — BEFORE ANY ELEMENT INTERACTION ***
(click | fill | hover | select | drag)
DO NOT GUESS SELECTORS FROM SCREENSHOTS.
Take a snapshot first to get real selectors and refs:
- Call
a11y_take-aria-snapshot,a11y_take-ax-tree-snapshot, orcontent_get-as-html(e.g. selector"form") to see actual ids, roles, and markup. a11y_take-aria-snapshotreturns a tree with refs (e1, e2, …) and stores them in context. In click, fill, hover, select, drag, scroll use the selector as:- a ref (
e1,@e1,ref=e1), - a Playwright-style expression (e.g.
getByRole('button', { name: 'Login' }),getByLabel('Email'),getByText('Register'),getByPlaceholder('…'),getByTitle('…'),getByAltText('…'),getByTestId('…')), or - a CSS selector. Refs are valid until the next snapshot or navigation; re-snapshot after page/DOM changes.
- a ref (
- Then use the returned selectors or refs for interaction. Guessing (e.g.
input[type="email"]) often fails because real markup may usetype="text", wrappers, or different attributes.
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 · 86 lines · 26 tokens per session scan A 8821db12b8ca
default is an agent published in the GitHub repository serkan-ozal/browser-devtools-claude (8 stars, last pushed 5mo ago), licensed MIT. It adds 26 tokens to every session and 1,242 once invoked, about $0.0001 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 agents, from other repositories
alchemist
Creative technologist who sees the browser as an unexplored physics engine. Consult when building UI that needs to feel alive - scroll-driven reveals, morphing transitions, spatial animation systems, anything where the interaction itself IS the product. Thinks in weight, tension, and breath before thinking in code.…
praman-sap-planner-cli
SAP UI5 test planner via Playwright CLI. Token-efficient alternative to MCP planner. Generates test plan + gold-standard spec using CLI commands.
mcp
The page outline agents read for line offsets, the in-browser WebMCP bridge and when to run it yourself, the rate limits, and how the surfaces are tested.
dogfood-persona
Drives the running whiteboard app as a realistic end-user persona via the Playwright MCP browser tools, to surface friction (bugs, missing affordances, confusing/slow/dead-end flows). Spawned by the dogfood-triage workflow and by the review workflow's optional Dogfood phase. Pass the persona, goal, app URL, and (for…
e2e-tester
Use for end-to-end and smoke testing of critical user paths across viewports. Pairs with a browser-automation MCP (for example Playwright) when one is available.
browser-tester
Focused functional browser tester. Tests ONLY what the prompt specifies — navigate, check, report pass/fail. MUST run in foreground — MCP tools do not work in background.