browser-core-efficient
/browser-core-efficient
Use when working in the embedded browser and you want the cheapest reliable interaction loop.
- Author
- holaboss-ai ↗
- Repository
- holaos ↗
- Licence
- Apache-2.0
- Last repository change
- 21.08.2026
Assessment and security
- Files in this package
- 1
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 21.08.2026, without references to retired models or APIs.
Apache-2.0, otwarta: można używać i zmieniać, z zachowaniem licencji i informacji o zmianach.
What it does and when to use it
Use when working in the embedded browser and you want the cheapest reliable interaction loop.
Original author description
Use when working in the embedded browser and you want the cheapest reliable interaction loop.
How it works
- Goals
- Default Loop
- Tool Selection
- State Discipline
- Wait Discipline
- Escalation Rules
SKILL.md preview
--- name: browser-core-efficient description: Use when working in the embedded browser and you want the cheapest reliable interaction loop. --- # Browser Core Efficient Use this skill when operating the workspace-embedded browser and efficiency matters. ## Goals - reduce browser tool round trips - reduce large state snapshots - reduce unnecessary screenshots and page text reads - prefer reliable one-call actions over browse-then-browse-again loops ## Default Loop 1. If the target is already known, use `browser_act` directly with locator signals. 2. If the target is semantically known but the page is large, use `browser_find`. 3. Use `browser_get_state` only for orientation or when the target is still unclear. 4. After page-changing actions, use `wait_for` on the action itself when possible. 5. Re-read state only if the action result is ambiguous. ## Tool Selection - Prefer `browser_act` over index-based tools when text, label, role, selector, or ref can identify the target. - Prefer `browser_find` before `browser_get_state` when you know what you want but not where it is. - Use `browser_click` and `browser_type` only when you already have a stable index from a fresh state snapshot. - Use `browser_select_tab` and `browser_close_tab` instead of spending extra calls re-orienting around tabs. - Use `browser_wait` for explicit stabilization, especially for URL changes, DOM changes, and download completion. ## State Discipline - Default to `browser_get_state` with compact detail. - Do not request page text unless the task is primarily about reading content. - Do not request screenshots unless DOM-first signals are insufficient. - When a fact may live in custom-element attributes, `data-*` attributes, `href`s, or hydration data rather than visible text, inspect that targeted page-local DOM source before concluding the fact is unavailable. - If a page change likely invalidated earlier refs or indexes, treat them as stale and re-locate the target. ## Wait Discipline - Prefer action-local `wait_for` over a separate follow-up wait call. - Use `interactive` or `domcontentloaded` after lightweight transitions. - Use full load completion only when the page truly needs it. - Use `download_started` and `download_completed` for browser-triggered downloads instead of manual polling through state reads. ## Escalation Rules - Escalate to `browser_get_state detail=standard` only when compact state is insufficient. - Escalate to `include_page_text=true` only when the page content itself is the task output. - Escalate to `browser_evaluate` for narrow read-only extraction when the required fact is likely present in page-local attributes or hydration data but not exposed cleanly through visible text. - Escalate to screenshots for visual ambiguity, layout validation, canvas/chart/PDF content, or explicit user-visible confirmation. - Escalate to debug or observability tools only after the cheaper path fails.
Versions
The database has no other skills named browser-core-efficient.
Identical content appears in 1 other GitHub folders. We show one copy with the highest score.
Database snapshot: 13.11.2026 · package version 6aba6790d4ba (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 95browser-control
Drive the user's existing Chromium-family browser with deterministic Playwright. Use when asked to inspect, automate, test, or interact with a visible browser tab; continue an authenticated browser workflow; handle 2FA, passkeys, CAPTCHAs, or payment confirmation; record browser… - 100browser
Drives a real browser from the shell: open, read, click, fill, sign in, download. Use it before the first browser command of any web job: a page, a form, a checkout, a sign-in, their own Chrome. - 81Chrome
Browser automation for the user's Chrome browser. Use for browser tasks that require the user's cookies, logged-in sessions, existing tabs, extensions, or remote authenticated sites. - 100agent-browser
Browser automation CLI for AI agents. Use this when asked to test something in a real browser. - 92ai-browser
Drive Chrome AI, the user's dedicated signed-in Chrome window, with the dev-browser CLI to open sites, read pages, click, fill forms, take screenshots and check a live app. Use when the user asks to browse, open a site, click, fill a form, take a screenshot, scrape a page, test… - 95browser
Automate web browser interactions using natural language via CLI commands. Use when the user asks to browse websites, navigate web pages, extract data from websites, take screenshots, fill forms, click buttons, or interact with web applications.