What is chrome-devtools-axi?

A command-line browser tool for operating Chrome and investigating live web pages. It can inspect what the browser currently renders, including the page structure, styles, console, network activity, screenshots, and responsive behavior.

Use it to browse, submit forms, extract rendered content, diagnose page and network problems, inspect layout changes, and investigate performance or memory behavior.

It removes the guesswork of answering questions about a page from static source code alone. It also provides observable results for browser actions such as filling forms or checking a failure.

chrome-devtools-axi — Skill for Claude Code and Codex by n-n-code

Install

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.

Any agent

npx skills add n-n-code/n-n-code-skills --skill chrome-devtools-axi

Clone the repo

git clone --depth 1 https://github.com/n-n-code/n-n-code-skills

Made for: Claude Code, Codex.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/n-n-code/n-n-code-skills/chrome-devtools-axi.svg)](https://agentmods.dev/skills/n-n-code/n-n-code-skills/chrome-devtools-axi)

Your own site

<a href="https://agentmods.dev/skills/n-n-code/n-n-code-skills/chrome-devtools-axi"><img src="https://agentmods.dev/badge/skills/n-n-code/n-n-code-skills/chrome-devtools-axi.svg" alt="Measured on agentmods" height="20"></a>

Per session 79 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,221 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00079	$0.01221
Opus 5	$0.00039	$0.00611
Sonnet 5	$0.00016	$0.00244
Haiku 4.5	$0.00008	$0.00122

Measured today against content hash 2d98a13f0208, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

chrome-devtools-axi 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 today.

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.

.agents/skills/chrome-devtools-axi/SKILL.md · 112 lines

How it starts

The opening of the file, as written. The whole thing — 112 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Chrome DevTools AXI

Answer the user's browser question with a reproducible observation, or complete the requested browser action and establish its result. This workflow needs an available shell, AXI and its runtime dependencies, and access to the selected target.

Determine what would count as completion

Identify the URL or existing tab, the question or action, and an observable success condition. Resolve available context before asking the user. Respect their chosen browser, tool, profile, and tab as well as any required host interface.

Select ownership by the requested work:

General Chrome operation and live investigation belong here.
A static response that needs no browser state can be fetched directly.
Explicit Playwright investigation and existing Playwright tests belong to playwright-testing; harness installation or repair belongs to setup-playwright.
UI changes or a broader UX review need the relevant UI skill. Security audits need security. Add tester-mindset when the investigation needs a test strategy, rather than making these skills routine prerequisites.

Use AXI as this package's execution interface. A missing executable is a capability gap to report, not a reason to switch to direct MCP automatically. An investigation request permits findings; source fixes, durable tests, and unrelated audits require that additional scope.

Carry out a routine session

Discover the executable's version and help. Choose an installed command when available. npx -y chrome-devtools-axi is an alternative only when package download/execution is authorized; use that same prefix throughout. Check unfamiliar subcommands through their own help.
Select a unique CHROME_DEVTOOLS_AXI_SESSION and task-owned isolated launch by default. Inspect inherited endpoint, profile, auto-connect, and port settings before the first browser action. Read operation and sessions for overrides, existing browsers, prerequisites, or recovery. Apply task settings again in each new shell invocation.
Establish the page's identity. List/select an existing target or open an owned page as appropriate, then verify its actual URL/title. Keep another person's tab intact when a new tab can serve the task.
Observe readiness and take a CLI snapshot before selecting a control. Locate it by meaning and current UID. Quote the complete returned reference, including its @g... prefix. For appearance, inspect a screenshot; for details outside the snapshot, evaluate only the relevant DOM/state.
Perform the authorized interaction using ordinary CLI commands. Keep page selection, snapshots, and actions sequential within a shared browser. Follow the existing task authority without asking again for routine steps.
Check the success condition using the resulting page, record, or request. Refresh the target and references after transitions. A completed command, an HTTP success code, or an empty error list does not establish the product outcome.
Verify any saved evidence, restore task-changed conditions in a reused browser, and close only pages or launch sessions owned by this task. Preserve a session the user wants left open and report meaningful residual state.

Read the full file on GitHub · 112 lines

Files

What ships with it

3 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.

Changes

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.

today First seen · 112 lines · 79 tokens per session scan A 2d98a13f0208

Subscribe to this mod's changes

chrome-devtools-axi is a skill published in the GitHub repository n-n-code/n-n-code-skills (4 stars, last pushed yesterday), licensed MIT. It adds 79 tokens to every session and 1,221 once invoked, about $0.0004 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 2026-09-06.

Browser automation Debugging agent-skills agents agentskills claude-skills claudeskills codex-skills

triage-issue

Verify an issue, persisted audit finding, or unresolved review finding against current code, classify it, and write only the contracted forge/ledger outcome. Supports independent batches and an immediate-fix review-finding route. Triggers: "triage-issue", "triage issue N", "triage this finding", "is this trigger met".

gtrabanco/agentic-workflow · 74 tokens

diagnosing-bugs

Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.

d0hx4n2503/Agent-Skills · 40 tokens

broken-link-checker

Scans a website to find broken links (404s, 500s). Crawls internal pages, identifies broken outbound links, and reports source pages for easy fixing. Use this when the user asks to "check for broken links", "find 404s", "audit my links", or "is my site healthy".

nowork-studio/notfair-plugin · 72 tokens

chrome-devtools-axi