Agent skill

Pandascript

by lightpanda-io in lightpanda-io/agent-skill

Write PandaScript agent scripts (.js) — Lightpanda's replayable browser-automation format, run token-free with lightpanda agent script.js.

Apache-2.0Auto-check passedProductivity & Automation

Install Pandascript

skills CLI
$ npx skills add lightpanda-io/agent-skill --skill pandascript -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install lightpanda-io/agent-skill pandascript --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/lightpanda-io/agent-skill.git skills-src && mkdir -p .claude/skills && cp -r skills-src/pandascript .claude/skills/pandascript && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
pandascript
GitHub stars
101
Token cost
~2.7k tokens
SKILL.md length
1,196 words
Files
1
Skills in repo
2
Repo updated
First seen
Licence
Apache-2.0

At a glance

Write PandaScript agent scripts (.js) — Lightpanda's replayable browser-automation format, run token-free with lightpanda agent script.js.

  • Works in 9 steps: Navigate, settle, read. After await… → List-to-detail — fan out independent… → evaluate is a last resort, not a reading… → …
  • Tasks that involve Browser automation
  • SKILL.md covers Mental model (get this right…, Primitives, extract schema and Best practices, plus 1 more section
  • Reaches news.ycombinator.com; needs LP_HN_PASSWORD

What it does

Pandascript is an agent skill from lightpanda-io/agent-skill. Write PandaScript agent scripts (.js) — Lightpanda's replayable browser-automation format, run token-free with lightpanda agent script.js.

Its SKILL.md is about 2.7k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Productivity & Automation, covering Browser automation. The repository describes itself as: Lightpanda browser skill for openclaw. The licence is Apache-2.0.

When your agent uses it

  • Tasks that involve Browser automation

Example prompts

  • “/pandascript”

Requirements

  • Node.js

Workflow steps

9 steps, taken from the first numbered list in SKILL.md.

  1. Navigate, settle, read. After await page.goto on a dynamic page (feeds, search results, comment threads), call…
  2. List-to-detail — fan out independent pages. Extract the list, then open one page per item and start every navigation together so the…
  3. evaluate is a last resort, not a reading tool. A querySelectorAll-and-parse page.evaluate block is always wrong: lift the raw strings with…
  4. Credentials via $LP_* placeholders in any string argument (page.fill("#pw", "$LP_HN_PASSWORD")). Never inline a real secret; placeholders…
  5. Unique selectors. Disambiguate with attributes/position: inputtype="submit", not input[type="submit"].
  6. Let failures fail. Primitives throw on error and stop the script — only try/catch where you have a real fallback (e.g. optional cookie…
  7. End with return . console.log is for debug output only and doesn't JSON-format objects.
  8. Modern, readable JS: const/let, for (const x of xs), template literals, destructuring, 2-space indent.
  9. Comment the intent of each block. Put a one-line // comment above each logical step describing what it accomplishes toward the goal (not…

What it can do on your machine

Read from SKILL.md and the folder at commit 387702e. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    No scripts in the folder and no shell commands in SKILL.md (its code samples are javascript and bash).

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    Hosts in commands or code, which the agent is likely to contact:

    • news.ycombinator.com

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names these keys or tokens, usually read from environment variables:

    • LP_HN_PASSWORD

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Pandascript loads about 2.7k tokens when it runs. Until then it costs about 38 tokens; SKILL.md has 1,196 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~38
When it runs · the whole SKILL.md, loaded when a task matches
~2.7k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from lightpanda-io/agent-skill at commit 387702e, republished under its Apache-2.0 licence (© lightpanda-io). 1,196 words, ~2,703 tokens.

Download SKILL.mdSave it as .claude/skills/pandascript/SKILL.md (or your agent's skills folder).
name
pandascript
description
Write PandaScript agent scripts (.js) — Lightpanda's replayable browser-automation format, run token-free with `lightpanda agent script.js`.

Writing Lightpanda agent scripts

Run with:

console
./lightpanda agent script.js

Mental model (get this right first)

The script runs in its own V8 context — neither the page nor Node.js:

  • Page is the only global. new Page() makes a page and await page.goto(url) navigates it; every other primitive is a method on that page: const page = new Page(); await page.goto(url); page.extract({...}); page.click(sel);.
  • No window, document, DOM, localStorage — read pages with page.extract(...), run page-side JS only via page.evaluate("...").
  • No require, process, fs, npm. Standard ECMAScript built-ins only (JSON, Map, template literals, …).
  • page.goto(...) is async — always await it. Page methods are synchronous: const data = page.extract({...}), never await page.extract(...). The script body runs as an async function, so top-level await is allowed.
  • Re-navigating reuses the same page: await page.goto(url2) keeps page valid and points it at the new URL, discarding the old page — read it before navigating away. Independent URLs don't share a page: make a new Page() for each and load them in parallel (fan-out, best practice 2).
  • Page evaluate("...") cannot see script variables — interpolate values into the string. Script code cannot see page variables.
  • Variables persist across navigations within one run, so cross-page aggregation is plain JS.
  • return <value> is the script's output, printed automatically (objects/arrays as JSON). End with return page.extract({...}); or return results;. A bare trailing expression is NOT printed; neither is console.log(JSON.stringify(...)).

Primitives

Page is the only global; new Page() makes a page and everything else is a method on it.

CallNotes
new Page()Makes a page object. No navigation yet — call page.goto(url) before any other method. Make several to navigate in parallel (fan-out, best practice 2).
page.close()Marks the page done; later method calls on it error. The page is otherwise reclaimed at script end.
await page.goto(url[, { timeout }])Async — must be awaited. Navigates the page (re-navigating reuses the same object). Waits for load. Rejects on navigation failure; a timeout does NOT reject (the page may still be usable). Default timeout 10000 ms.
page.evaluate(script[, { url, timeout, save }])Page-side JS escape hatch; returns text (JSON for objects/arrays).
page.extract(schema)The only primitive returning a real JS value (object/array). The schema is its only argument. See schema below.
page.click(selector)
page.fill(selector, value)
page.scroll([{ x, y }])
page.waitForSelector(selector[, { timeout }])waitFor* default timeout 5000 ms.
page.waitForScript(script[, { timeout }])Re-evaluates page JS until truthy.
page.waitForState(state[, { timeout }])state: one of "load", "domcontentloaded", "networkalmostidle", "networkidle", "done".
page.hover(selector)
page.press(selector, key)Selector first! page.press("Enter") binds "Enter" to selector and fails — use page.press(null, "Enter") or page.press({ key: "Enter" }).
page.selectOption(selector, value)
page.setChecked(selector[, checked])checked defaults to true.

Options (the trailing { … } object; every option may be omitted):

  • goto: timeout — Optional timeout in milliseconds. Defaults to 10000.
  • evaluate: url — Optional URL to navigate to before evaluating. timeout — Optional timeout in milliseconds. Defaults to 10000. save — Optional bridge-store key. The evaluate's return value is stored under this name and re-exposed as lp.<name> to subsequent evaluates. Objects, arrays, and strings are serialized automatically — no JSON.stringify needed.
  • scroll: x — Optional: The horizontal scroll offset. y — Optional: The vertical scroll offset.
  • waitForSelector: timeout — Optional timeout in milliseconds. Defaults to 5000.
  • waitForScript: timeout — Optional timeout in milliseconds. Defaults to 5000.
  • waitForState: timeout — Optional timeout in milliseconds. Defaults to 5000.

Calling convention: leading positionals + optional trailing options object, or one object with everything (page.waitForSelector("#row", { timeout: 2000 }) ≡ page.waitForSelector({ selector: "#row", timeout: 2000 })). A bare option positional (page.waitForSelector("#row", 2000)) and a field passed both ways are invalid arguments. null skips a positional. Arguments must be JSON-serializable.

CSS selectors only — backendNodeIds don't exist here. Standard CSS only: no jQuery :contains() or Playwright :has-text().

extract schema

Keys = output field names; values pick what to lift (not a JSON Schema):

js
const { stories } = page.extract({
  stories: [{
    selector: "tr.athing",          // one record per match
    limit: 5,
    fields: {                        // resolved relative to each match
      title: ".titleline > a",      // first match's text (null if missing)
      url: { selector: ".titleline > a", attr: "href" },
      text: ""                       // "" = the matched element's own text
    }
  }]
});
  • "sel" → first match's text; ["sel"] → all matches' text; { selector, attr } / [{ selector, attr }] → attribute(s); limit: N caps any array form.
  • Every value is a string or null — parse numbers in script logic.
  • Empty arrays are valid results; if every field misses, extract throws ("no schema selector matched any element") → your selectors are wrong, not the page empty.
  • An object schema always returns an object (destructure it); a bare array schema returns the array directly.
  • No save: option in scripts — keep results in variables.
Show full SKILL.md (618 more words)Show less

Best practices

  1. Navigate, settle, read. After await page.goto on a dynamic page (feeds, search results, comment threads), call page.waitForState("networkidle") or page.waitForSelector(...) before extracting. Most static pages are complete at load — don't wait blindly.
  2. List-to-detail — fan out independent pages. Extract the list, then open one page per item and start every navigation together so the detail pages load in parallel instead of one-after-another:
    js
    const list = new Page();
    await list.goto(listUrl);
    const { items } = list.extract({ items: [{ selector: "a.row", fields: { url: { attr: "href" } } }] });
    
    const pages = items.map(() => new Page());
    await Promise.all(pages.map((p, i) => p.goto(items[i].url)));   // all in flight at once
    return pages.map((p, i) => ({ ...items[i], ...p.extract({ /* schema */ }) }));
    • Concurrency is bounded by the HTTP connection pool: 40 total (--http-max-concurrent) and 6 per host (--http-max-host-open, the browser default — raising it much higher risks overwhelming the target server). Extra navigations queue rather than fail, so a same-site fan-out loads ~6 pages at a time. For long lists, fan out in batches and page.close() each page once read so its memory is reclaimed.
    • Promise.all rejects the whole batch if any goto fails (a timeout does not reject); use Promise.allSettled when partial results are fine.
    • Walk serially on one page (for (const it of items) { await page.goto(it.url); … }) only when the steps depend on each other — each page decides the next URL, or they share login/session state.
  3. evaluate is a last resort, not a reading tool. A querySelectorAll-and-parse page.evaluate block is always wrong: lift the raw strings with page.extract, then trim/split/parse them in top-level JS. Reserve page.evaluate for behavior that must run inside the page and no builtin covers — and remember its state dies on every navigation/reload, while script variables persist.
  4. Credentials via $LP_* placeholders in any string argument (page.fill("#pw", "$LP_HN_PASSWORD")). Never inline a real secret; placeholders resolve inside the Lightpanda process.
  5. Unique selectors. Disambiguate with attributes/position: input[type="submit"][value="login"], not input[type="submit"].
  6. Let failures fail. Primitives throw on error and stop the script — only try/catch where you have a real fallback (e.g. optional cookie banner: try { page.click("#accept") } catch {}).
  7. End with return <result>. console.log is for debug output only and doesn't JSON-format objects.
  8. Modern, readable JS: const/let, for (const x of xs), template literals, destructuring, 2-space indent.
  9. Comment the intent of each block. Put a one-line // comment above each logical step describing what it accomplishes toward the goal (not restating the call). One comment per block, not per line — skip self-evident lines:
    js
    // Load the Hacker News front page
    const page = new Page();
    await page.goto("https://news.ycombinator.com");
    
    // Pull the top 5 stories (title + link)
    const { stories } = page.extract({ stories: [{ selector: "tr.athing", limit: 5, fields: { title: ".titleline > a", url: { selector: ".titleline > a", attr: "href" } } }] });
    
    // Open each story page in parallel and read its text
    const pages = stories.map(() => new Page());
    await Promise.all(pages.map((p, i) => p.goto(stories[i].url)));

Common errors

ErrorCause / fix
extract is not defined (or click/fill/…)These are methods on the page object, not globals → const page = new Page(); await page.goto(url); page.extract(...)
Page must be called with newPage(...) called without new → const page = new Page();
page is not navigated or has been closedA method on a fresh new Page() (or a closed page) → await page.goto(url) first
page handle is no longer validUsed a page after a later goto on the same page replaced it → read it before navigating away. Sibling pages from other new Page() calls stay valid.
document is not definedDOM API in script context → use page.extract or page.evaluate
require is not definedNot Node.js
no page loaded - run page.goto(url) firstPage method before navigation
invalid argumentsWrong arity/shape, non-JSON value, or a field set both positionally and in options
extract: no schema selector matched any elementAll schema fields missed → fix selectors
press fails with one string argSelector-first: use page.press(null, "Enter") or page.press({ key: "Enter" })

© lightpanda-io, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in pandascript of lightpanda-io/agent-skill.

Open the folder on GitHubat commit 387702e

Compare with similar skills

Pandascript next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Pandascript compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Pandascript this skilllightpanda-io/agent-skill101—~2.7kAutomated safety check: PassApache-2.0
Agent Browserquran/quran.com-frontend-next1.9k42 repos~3.3kAutomated safety check: PassNone
Dev-Browser CLI AutomationSawyerHood/dev-browser6.7k1 repos~455Automated safety check: PassMIT
Agent Browsersuperagent-ai/grok-cli3.5k1 repos~633Automated safety check: PassMIT
Browser Automationopenclaw/openclaw392k—~2.9kAutomated safety check: PassMIT
Camoufox CLIBin-Huang/camoufox-cli3501 repos~4.5kAutomated safety check: PassMIT

Similar skills

  • Agent Browser

    quran/quran.com-frontend-next

    Automates browser interactions for web testing, form filling, screenshots, and data extraction.

    1.9k GitHub starsUsed in 42 repos~3.3k tokens
    Productivity & AutomationAuto-check passed
  • Dev-Browser CLI Automation

    SawyerHood/dev-browser

    Browser automation with persistent named pages via the dev-browser CLI. Use when users ask to navigate websites, fill forms, take screenshots, extract web…

    6.7k GitHub starsUsed in 1 repo~455 tokens
    Productivity & AutomationAuto-check passed
  • Agent Browser

    superagent-ai/grok-cli

    Use the host-side agent-browser CLI for local browser smoke tests, screenshots, snapshots, and simple UI validation against forwarded localhost URLs.

    3.5k GitHub starsUsed in 1 repo~633 tokens
    Productivity & AutomationAuto-check passed
  • Browser Automation

    openclaw/openclaw

    A skill your agent uses when controlling web pages with the OpenClaw browser tool, especially multi-step flows, login checks, tab management, or recovery from stale refs/timeouts.

    392k GitHub stars~2.9k tokensUpdated today
    Productivity & AutomationAuto-check passed
  • Camoufox CLI

    Bin-Huang/camoufox-cli

    Anti-detect browser automation CLI & Skills for AI agents. An agent skill from Bin-Huang/camoufox-cli.

    350 GitHub starsUsed in 1 repo~4.5k tokens
    Productivity & AutomationAuto-check passed
  • Browser

    VibiumDev/vibium

    Automate browsers with the Vibium CLI. An agent skill from VibiumDev/vibium.

    2.9k GitHub stars~4.8k tokensUpdated today
    Productivity & AutomationAuto-check passed

More from lightpanda-io/agent-skill

  • Lightpanda

    lightpanda-io/agent-skill

    Lightpanda browser, drop-in replacement for Chrome-based browsing in any AI agent - faster and lighter for tasks without graphical rendering like data retrieval.

    101 GitHub stars~6k tokensUpdated yesterday
    Auto-check passed

Questions about Pandascript

What does Pandascript do?

Write PandaScript agent scripts (.js) — Lightpanda's replayable browser-automation format, run token-free with lightpanda agent script.js. Pandascript is an agent skill from lightpanda-io/agent-skill.js.

When should I use Pandascript?

Pandascript fits situations like: tasks that involve Browser automation.

How do I install Pandascript in Claude Code?

Run `npx skills add lightpanda-io/agent-skill --skill pandascript -a claude-code`. Or copy the skill folder (pandascript in lightpanda-io/agent-skill) into .claude/skills/pandascript in your project. Claude Code loads it when a task matches its description.

How do I install Pandascript in Codex?

Run `npx skills add lightpanda-io/agent-skill --skill pandascript -a codex`. Or copy the skill folder (pandascript in lightpanda-io/agent-skill) into .agents/skills/pandascript in your project. Codex loads it when a task matches its description.

Can I use Pandascript in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add lightpanda-io/agent-skill --skill pandascript -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/pandascript, .gemini/skills/pandascript, .github/skills/pandascript and .opencode/skills/pandascript in your project.

What does Pandascript need to run?

Going by SKILL.md and its folder, Pandascript needs credentials named LP_HN_PASSWORD. Our summary lists: Node.js.

Does Pandascript access the network?

SKILL.md names 1 domain. In commands or code: news.ycombinator.com; the agent is likely to contact it when it follows the instructions. This is read from the text; nothing was executed.

Is Pandascript safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Pandascript use?

Pandascript is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Pandascript use?

About 2.7k tokens (SKILL.md is roughly 11k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Pandascript?

Skills that share tags, products or a category with Pandascript: Agent Browser (quran/quran.com-frontend-next, 1.9k stars), Dev-Browser CLI Automation (SawyerHood/dev-browser, 6.7k stars), Agent Browser (superagent-ai/grok-cli, 3.5k stars) and Browser Automation (openclaw/openclaw, 392k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Pandascript?

lightpanda-io (a GitHub organization) maintains it in lightpanda-io/agent-skill, which has 101 GitHub stars. The repository holds 2 skills in this directory. The repository was last updated on October 6, 2026.

Source: lightpanda-io/agent-skill on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.