web

web is a skill for Claude Code, Codex from xuzhougeng/wispterm. It costs 9 tokens per session (1,443 once invoked), scanned A, original, MIT.

A browser skill for inspecting and interacting with live web pages. It can work with an existing browser session and focuses on checking visible page content and changes after actions.

In plain words
What is it for?
Use it to read pages, inspect page structure, fill forms, click controls, and verify navigation, dialogs, notifications, or validation errors.
Why use it?
It helps an agent navigate real websites while keeping track of what changed after each click or form submission. This reduces mistakes caused by relying on stale or incomplete page information.

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/xuzhougeng/wispterm/browser
Any agent
npx skills add xuzhougeng/wispterm --skill browser
Clone the repo
git clone --depth 1 https://github.com/xuzhougeng/wispterm

Made for: Claude Code, Codex.

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.

agentmods badge for web

README.md
[![agentmods](https://agentmods.dev/badge/skills/xuzhougeng/wispterm/browser.svg)](https://agentmods.dev/skills/xuzhougeng/wispterm/browser)
Your own site
<a href="https://agentmods.dev/skills/xuzhougeng/wispterm/browser"><img src="https://agentmods.dev/badge/skills/xuzhougeng/wispterm/browser.svg" alt="Measured on agentmods" height="20"></a>
Per session 9 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 1,443 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 $0.00009 $0.01443
Opus 5 $0.00005 $0.00722
Sonnet 5 $0.00002 $0.00289
Haiku 4.5 $0.00001 $0.00144

Measured 5d ago against content hash 0c9959286959, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

web 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 5d 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.

tests/eval/skills/browser/SKILL.md · 167 lines

How it starts

The opening of the file, as written. The whole thing — 167 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Web

Use this skill when the user asks to inspect, navigate, extract from, or act on live web pages. Prefer read-only inspection unless the user explicitly asks to interact with the page or submit data.

GenericAgent-inspired browser workflow

This default workflow is adapted for WispTerm from GenericAgent's browser-agent pattern: keep the browser state real, keep observations compact, use precise page execution when available, and verify every page-changing action.

  1. Use the user's existing browser/session context when the available tools support it. Logged-in state is often more valuable than a clean sandbox.
  2. Capture a compact text or DOM snapshot before acting. Prefer the main page content, active dialog, focused form, and visible controls over full raw HTML.
  3. For interaction, prefer precise selectors, small JavaScript snippets, or browser-native automation over broad coordinate guesses. If a click or form change is sensitive, assume synthetic events may be rejected and verify the result.
  4. After every action, inspect the delta: changed text, changed DOM region, navigation, reload, new tab, toast, or validation error.
  5. Avoid repeated full-page scans. Reuse tab identity, URLs, visible labels, and earlier observations; rescan only after navigation or meaningful page change.
  6. If the browser tool layer cannot reach a page, iframe, upload control, or trusted event path, state the limitation and switch to an available fallback such as terminal HTTP requests, local scripts, or user confirmation.

Reusable JavaScript snippets

Use these snippets with the available browser JavaScript execution tool when that tool exists. They are intentionally compact: paste the helper into the same execution as the action that needs it, then return JSON-shaped data.

Compact page snapshot

function wisptermCompactSnapshot(limit = 120) {
  const text = (node) => (node?.innerText || node?.textContent || "")
    .replace(/\s+/g, " ")
    .trim();
  const isVisible = (el) => {
    const rect = el.getBoundingClientRect();
    const style = getComputedStyle(el);
    return rect.width > 1 &&
      rect.height > 1 &&
      style.display !== "none" &&
      style.visibility !== "hidden" &&
      Number(style.opacity || 1) > 0;
  };
  const short = (value, max = 180) => {
    value = String(value || "").replace(/\s+/g, " ").trim();
    return value.length > max ? value.slice(0, max) + " ..." : value;
  };
  const describe = (el) => {
    const rect = el.getBoundingClientRect();
    return {
      tag: el.tagName.toLowerCase(),
      id: el.id || undefined,
      name: el.getAttribute("name") || undefined,
      type: el.getAttribute("type") || undefined,
      role: el.getAttribute("role") || undefined,
      aria: el.getAttribute("aria-label") || undefined,
      placeholder: el.getAttribute("placeholder") || undefined,
      value: /^(input|textarea|select)$/i.test(el.tagName) ? short(el.value, 80) : undefined,
      href: el.href ? short(el.href, 160) : undefined,
      label: short(el.getAttribute("aria-label") || el.getAttribute("title") || text(el), 140),
      rect: {
        x: Math.round(rect.x),
        y: Math.round(rect.y),
        w: Math.round(rect.width),
        h: Math.round(rect.height)
      }
    };
  };
  const controlSelector = [
    "dialog",
    "[role='dialog']",
    "[aria-modal='true']",
    "button",
    "a[href]",
    "input:not([type='hidden'])",
    "textarea",
    "select",
    "[role='button']",
    "[role='menuitem']",
    "[contenteditable='true']"
  ].join(",");
  const controls = Array.from(document.querySelectorAll(controlSelector))
    .filter(isVisible)
    .slice(0, limit)
    .map(describe);
  const headings = Array.from(document.querySelectorAll("h1,h2,h3,[role='heading']"))
    .filter(isVisible)
    .slice(0, 40)
    .map((el) => short(text(el), 140))
    .filter(Boolean);
  return {
    url: location.href,
    title: document.title,
    active: document.activeElement ? describe(document.activeElement) : null,
    headings,
    controls,
    bodyText: short(text(document.body), 4000)
  };
}
return wisptermCompactSnapshot();

Read the full file on GitHub · 167 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.

  1. 5d ago First seen · 167 lines · 9 tokens per session scan A 0c9959286959

Subscribe to this mod's changes

web is a skill published in the GitHub repository xuzhougeng/wispterm (401 stars, last pushed today), licensed MIT. It adds 9 tokens to every session and 1,443 once invoked, about $0.0000 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

computer-use

Use Orca's computer-use CLI for OS/window-level inspection and input in visible local app windows. Use when a task must read or operate a native app or an external browser window (for example, Chrome, Edge, or Safari) or an app webview. Do not use for Orca's embedded browser or page-only browser automation. Use…

stablyai/orca · 99 tokens

liney-cli

Use the Liney CLI (liney) to inspect or control a running Liney terminal workspace and the agent sessions it hosts. Liney runs several coding agents in parallel, each in its own pane/tab/worktree, so reach for this whenever the user wants to act on a pane other than the current one — check on, coordinate, read from…

everettjf/liney · 195 tokens

terminal-setup-install

Idempotent macOS terminal installer for Ghostty, Oh My Zsh, Powerlevel10k, Glow, MesloLGS Nerd Font, plus optional markdown preview and clickable-path extras.

YoungLeadersDotTech/young-leaders-tech-marketplace · 44 tokens

linear-tickets

Use Orca's Linear CLI through orca linear ... commands to read linked ticket context with orca linear issue --current --full --json, post completion updates, move work forward through Linear workflow states, attach PR/MR links with orca linear attach --current --url --title "PR/MR link" --json, and triage Linear tasks…

stablyai/orca · 160 tokens

orca-cli

Use the public orca CLI to operate Orca-managed worktrees, folder contexts, terminals, repos, automations, artifacts, skill sharing, worktree comments, and the browser embedded inside the Orca app. Use when the user says "$orca-cli", "use orca cli", "Orca worktree", "child worktree", "cardStatus", "spawn codex/claude…

stablyai/orca · 249 tokens

orca-emulator-android

Control an Android emulator / device from inside Orca using the orca CLI. Use for listing/booting AVDs, taps, swipes, typing, hardware buttons (incl. Back and Recents), rotation, app install/launch, runtime permissions, the accessibility tree, and logcat — driving a real adb-connected device or emulator.…

stablyai/orca · 104 tokens