fuse-browser-usage
Use when about to call any mcp__fuse-browser__* tool. Routes fetch/crawl/SERP vs live browser session vs screenshot capture, and enforces the 4 ZERO-TOLERANCE rules -- read BEFORE the tool call.
SKILL.mdplugins/ai-pilot/skills/fuse-browser-usage/SKILL.md
GitHub<objective>
This is the canonical doctrine for every `mcp__fuse-browser__*` call: 4 zero-tolerance rules (fast-path first with no browser launch for reading, one session always closed, batch over per-URL loops, deterministic schema-based extraction over manual snapshot parsing), then routing to one of three reference profiles depending on the goal -- research/docs (fetch, crawl, SERP, no pixels), webapp testing (console, network, live interactions), or visual/design (screenshots, responsive + dark mode, regression diffing).
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.
</objective>
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)
- Fast-path FIRST (no browser launch, ~10× faster) —
browser_fetch/browser_fetch_batch/browser_crawl/browser_serp_batchto 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. - One session, always closed —
browser_openonce → reuse thesessionIdacrossbrowser_navigatecalls →browser_closewhen done. Never leak sessions; the 2-min TTL is a safety net, not a substitute for closing. - Batch over loops —
browser_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. - Deterministic extraction —
browser_extract_schemawithcontainerSelector(card-by-card, correlated fields) over manualbrowser_snapshotparsing.
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 |
- No live session for reading — if you only need text/HTML/data, the answer is always fast-path (profile 1). Never
browser_opento read. - Validation tools (any profile):
browser_visual_diff(regression vs baseline),browser_metrics(real Core Web Vitals),browser_console/browser_network/browser_cookies(runtime + security checks). - Full design pipeline is NOT here — for identity → generate → audit, use
fuse-design:design-webthen thefuse-design:design-reviewgate. This skill only covers the raw browser tooling.