A command for creating, developing, and building browser extensions with extension.dev, a platform for browser-extension projects.

Scaffolding a new extension, starting development with live reloading, and creating production builds for Chrome or Firefox.

It turns common extension tasks into guided commands with the project’s expected development and build steps.

extension — Command for Claude Code by extensiondev/mcp

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.

Clone the repo

git clone --depth 1 https://github.com/extensiondev/mcp

Made for: Claude Code.

Or install extension-mcp, the plugin that ships this one along with the rest of its 4 commands, 1 MCP server.

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/commands/extensiondev/mcp/extension/github.svg)](https://agentmods.dev/commands/extensiondev/mcp/extension)

Your own site

<a href="https://agentmods.dev/commands/extensiondev/mcp/extension"><img src="https://agentmods.dev/badge/commands/extensiondev/mcp/extension/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/commands/extensiondev/mcp/extension"><img src="https://agentmods.dev/badge/commands/extensiondev/mcp/extension.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 12 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 1,599 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 1 finding. 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	$0.00012	$0.01599
Opus 5.5	$0.00005	$0.00640
Sonnet 5.5	$0.00002	$0.00320
Haiku 4.5	$0.00001	$0.00160

Measured 7d ago against content hash 94b6e23bcdc3, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

extension scanned grade A with 1 finding 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 7d 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

2. If not, check the template catalog: `curl -sL "https://registry.extension.land$(curl -sL https://registry.extension.land/extension-dev/catalog/latest.json | jq -r .catalog)" | jq '.templates[] | {slug, description, ui

claude/commands/extension.md · 90 lines

How it starts

The opening of the file, as written. The whole thing — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are helping build a browser extension with the extension.dev platform. The user said: $ARGUMENTS

Actions

Parse the user's intent from $ARGUMENTS and execute the matching action:

"create " or "new ", Scaffold a new extension

If MCP tool extension_templates is available, use it to find the best template matching the user's description (check for surface type, framework, and keywords)
If not, check the template catalog: curl -sL "https://registry.extension.land$(curl -sL https://registry.extension.land/extension-dev/catalog/latest.json | jq -r .catalog)" | jq '.templates[] | {slug, description, uiFramework, surfaces, downloads}'
If MCP tool extension_create is available, use it (its answer's nextSteps is the next step); otherwise run npx extension@latest create <name> --template=<best-match>
Report what was created. A scaffold needs no build to verify it: do not run npm run build to check it; the next step is running it (below)

"dev" or "run", Start development

If MCP tool extension_dev is available, call it with the project path: it starts the dev server and the browser, waits for the session and answers with ready and firstLogs. ready.status: "ready" means the session is up: do not call extension_wait after it (that tool is for sessions started with wait: false, an answer whose ready.status was not ready, and build-only sessions)
Otherwise run npm run dev (or npx extension dev) yourself in the project root; never hand the command to the user as their next step
Tell the user the browser opened with their extension loaded
Mention HMR is active, changes will hot-reload

"build", Build for production

If MCP tool extension_build is available, use it (chrome by default; its answer carries outputPath). Otherwise run npm run build (or npx extension build) yourself
After success, report the output folder the tool or the CLI answered with, never an assumed one: a plain npm run build passes no --browser, so the engine targets chromium and writes dist/chromium/; --browser=chrome and extension_build write dist/chrome/
If the user mentions "firefox" or "both", also build with --browser=firefox
If they mention "zip" or "store", add --zip
To say what is in the build (sizes, entry points, store checks), call extension_analyze with the same project path and browser; do not list or read the output folder with shell commands

Read the full file on GitHub · 90 lines

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.

7d ago Changed 94b6e23bcdc3
9d ago Changed 8d64e155f9df
10d ago Changed · +11 lines 19f770cf2bbe
18d ago Changed · +2 lines d54f3bcb2656
22d ago Changed 5da8d098a243
1mo ago First seen · 77 lines · 12 tokens per session scan A 245208730631

Subscribe to this mod's changes

extension is a command published in the GitHub repository extensiondev/mcp (0 stars, last pushed today), licensed Apache-2.0. It adds 12 tokens to every session and 1,599 once invoked, about $0.0000 per session on Opus 5.5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Browser automation Claude Code agents ai-agent browser-extension chrome-extension claude

agentganggang

Use MultiAiPanel when you want Claude Code to compare one prompt across multiple supported AI chat tabs through the local MultiAiPanel MCP surface.

xiaojiou176-open/multi-ai-panel · 0 tokens

expect

Diff-aware AI browser testing — reads the git diff, maps changes to affected pages via the route map, generates a targeted test plan, and executes it via agent-browser (Rust daemon + CDP, ARIA-tree-first) with pass/fail reporting. Use when testing UI changes, verifying PRs before merge, or running regression checks on…

yonatangross/orchestkit · 73 tokens

page-serve

Hand a human a rendered HTML page at a stable HTTPS URL. Registers a portless route for a file or directory (https://NAME.localhost/, never a :port), prints the URL, optionally screenshots it, and tears it down on stop. Use when an agent produced a page a human is meant to open: glyph explainers, playgrounds, decision…

yonatangross/orchestkit · 96 tokens

extension