fuse-browser-usage

A usage guide for browser tools that fetch pages, search results, crawl websites, open interactive browser sessions, and capture screenshots. It explains which browser method fits each task and how to manage sessions.

In plain words
What is it for?
Use it before calling browser tools to read pages, scrape search results, interact with JavaScript sites, or capture responsive screenshots.
Why use it?
It helps avoid opening a live browser when a faster page fetch is enough and prevents browser sessions from being left open.

Skill for Claude CodeCodex

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.

agentmods
npx agentmods add skills/fusengine/agents/fuse-browser-usage
Any agent
npx skills add fusengine/agents --skill fuse-browser-usage
Clone the repo
git clone --depth 1 https://github.com/fusengine/agents

Made for: Claude Code, Codex.

Per session 55 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 819 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00055 $0.00819
Opus 5 $0.00028 $0.00409
Sonnet 5 $0.00011 $0.00164
Haiku 4.5 $0.00006 $0.00082

Measured 2d ago against content hash f3c4f6a4a205, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

fuse-browser-usage 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.

plugins/ai-pilot/skills/fuse-browser-usage/SKILL.md · 45 lines

How it starts

The opening of the file, as written. The whole thing — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.

It covers only the raw browser tooling. The full design pipeline (identity -> generate -> audit) is NOT here -- that lives in fuse-design:design-web and the fuse-design:design-review gate.

fuse-browser — Efficient Usage

Canonical doctrine for every mcp__fuse-browser__* call. Read this skill FIRST, then load the one reference matching your goal. Default engine is patchright (stealth identity auto-generated); the Chromium binary comes from setup.sh.

4 Rules (ZERO TOLERANCE)

  1. Fast-path FIRST (no browser launch, ~10× faster)browser_fetch / browser_fetch_batch / browser_crawl / browser_serp_batch to read a page, bulk-fetch, crawl a site, or scrape Google SERP. Open a live session ONLY when you need interaction, JS rendering, or pixels.
  2. One session, always closedbrowser_open once → reuse the sessionId across browser_navigate calls → browser_close when done. Never leak sessions; the 2-min TTL is a safety net, not a substitute for closing.
  3. Batch over loopsbrowser_serp_batch (N queries), browser_fetch_batch / browser_shots_batch (N URLs), browser_screenshot { viewports:[...], colorScheme } (responsive + dark in ONE call). Never write a per-URL loop.
  4. Deterministic extractionbrowser_extract_schema with containerSelector (card-by-card, correlated fields) over manual browser_snapshot parsing.

Routing — pick the profile, load the reference

Goal Profile Reference
Read docs, bulk-fetch, crawl, scrape SERP (no pixels) Research / docs references/research-docs.md
Test a running webapp (console, network, interactions) Webapp testing references/webapp-testing.md
Screenshots, responsive + dark, visual regression Visual / design references/visual-design.md

Read the full file on GitHub · 45 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.

  1. 2d ago First seen · 45 lines · 55 tokens per session scan A f3c4f6a4a205

Subscribe to this mod's changes

fuse-browser-usage is a skill published in the GitHub repository fusengine/agents (25 stars, last pushed 29d ago), licensed MIT. It adds 55 tokens to every session and 819 once invoked, about $0.0003 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.

Related

Other skills, from other repositories