pagerunner-skill（寻呼员技能）

Pagerunner Skill — Quick Start Guide

Pagerunner is real Chrome browser automation for AI agents. It gives Claude, Cursor, Windsurf, or any MCP client native control over your real Chrome — with your existing login sessions, cookies, and browser history already loaded.

For AI Agents: Decision Guide

If you are an AI agent executing a task with this skill, follow these rules. They are non-negotiable — they prevent the most common failures.

Rule 1: Every workflow starts the same way

const sessionId = await open_session({ profile: "personal" });  // or "agent-work", etc.
const [tab] = await list_tabs(sessionId);
const tabId = tab.target_id;  // Save this — every tool needs it

Get target_id from list_tabs before anything else. Without it, no other tools work.

Rule 2: wait_for before every interaction (no exceptions)

After navigate(), the page is loading. Selectors don't exist yet. Always wait before clicking, filling, or evaluating:

await navigate(sessionId, tabId, url);
await wait_for(sessionId, tabId, { selector: ".any-stable-element", ms: 5000 });
// Only now is it safe to click, fill, get_content, evaluate

Skipping this causes "selector not found" errors. If you don't know a stable selector, use get_content first to find one.

Rule 3: Understand the page before acting

Don't guess at selectors. Call get_content to see what's on the page, then act:

const structure = await get_content(sessionId, tabId);
// Reveals: visible text, form fields, buttons, navigation state
// Use this to find the right selectors before fill/click/evaluate

Rule 4: Which input tool to use

Rule 5: Verify after every action

After submitting a form or clicking a button, confirm it worked:

await click(sessionId, tabId, ".submit-btn");
await wait_for(sessionId, tabId, { selector: ".success-message, .error-message", ms: 5000 });
const result = await get_content(sessionId, tabId);
// Check result contains expected confirmation — never assume success

Rule 6: evaluate() must return labeled objects

// ❌ Never — array field order is ambiguous, causes wrong answers
return [likes, replies];

// ✅ Always — labeled fields are unambiguous
return { likes, replies };

Pagerunner's metadata block will warn you if you return an array. Read it.

Rule 7: Always close sessions with try/finally

const sessionId = await open_session({ profile: "..." });
try {
  // ... workflow ...
} finally {
  await close_session(sessionId);  // Runs even if the workflow throws
}

close_session also writes an auto-checkpoint (v0.6.0+), preserving session state.

Choose Your Path

Solo Developer (Claude Code / Cursor)

Goal: Close the implementation loop. Edit code → see the result in the browser → iterate without manual verification.

Quick Start (5 lines):

const sessionId = await open_session({ profile: "personal" });
const [tab] = await list_tabs(sessionId);
await navigate(sessionId, tab.target_id, "http://localhost:3000");
await screenshot(sessionId, tab.target_id);  // see what you built
await close_session(sessionId);

Killer Feature: Your Chrome, already logged into everything. No API token setup.

Learn More: See PATTERNS.md → "Frontend dev loop"

📱 Power User (OpenClaw / Hermes)

Goal: Get browser tasks done from your phone while the laptop runs unattended.

Quick Start (8 lines):

// First time: log in manually and save the session
pagerunner open-session agent-work
// ... perform login steps ...
pagerunner save-snapshot <session> <tab> https://jira.mycompany.com

// Later: agent profile is pre-authenticated
pagerunner open-session agent-work
pagerunner restore-snapshot <session> <tab> https://jira.mycompany.com
// Agent is logged in. Now do work: navigate, get_content, fill forms

Killer Features: Agent profile isolation + snapshot persistence + daemon for always-on

Real Example:

WhatsApp: "Check my Jira for blockers"
→ OpenClaw triggers skill
→ open_session(profile="agent-work") → restore_snapshot
→ navigate to Jira → get_content → summarize
→ screenshot → send back to WhatsApp

Learn More: See PATTERNS.md → "Authentication persistence"

🔒 Security-Conscious (NemoClaw / regulated industries)

Goal: Automate workflows on sensitive data without PII reaching the LLM.

Quick Start (1 flag):

open_session({
  profile: "agent",
  anonymize: true  // That's it. PII stripped before it reaches you.
});

// Every get_content and evaluate result has PII replaced with tokens:
// john@company.com → [EMAIL:a3f9b2]
// Claude works with tokens
// Pagerunner de-tokenizes only when writing to forms

// Audit log records every action (compliance proof)

Killer Features: PII never leaves your machine in plaintext + local NER + audit trail

Learn More: See SECURITY.md → "Anonymization modes"

⚙️ Server-Side / Infrastructure (Hermes + cron)

Goal: Persistent browser automation across scheduled runs. Agents coordinate via shared state.

Quick Start (daemon setup):

pagerunner daemon &  # Run once, holds the DB lock

# Now use Pagerunner in cron, shell scripts, or Hermes tasks
# Multiple agents share state via KV store

# v0.6.0: Chrome windows SURVIVE daemon restarts
kill $(pgrep -f "pagerunner daemon") && pagerunner daemon &
# Sessions auto-reattach — no lost work

// Agent A: collects data
await kv_set("pipeline", "pricing_data", JSON.stringify(results));

// Agent B (later): continues where A left off
const data = JSON.parse(await kv_get("pipeline", "pricing_data"));

Killer Features: Daemon + KV coordination + snapshots for auth handoff + session persistence across restarts (v0.6.0+)

Learn More: See ADVANCED.md → "Session Persistence & Auto-Reattach"

Common Gotchas

1️⃣ Arrays Cause Hallucinations

Problem: evaluate() returns [25, 2]. Claude guesses "25 likes, 2 replies" but it's actually "25 views, 2 likes."

Why: Arrays have no field labels. Order is ambiguous.

Solution: Always return labeled objects from evaluate():

// ❌ BAD
return [likes, replies];

// ✅ GOOD
return { likes, replies };

Pro Tip: Pagerunner metadata warns you if evaluate returns an array. Read the metadata block.

2️⃣ Selectors Timing

Problem: React/Vue renders asynchronously. Selector might not exist for 500ms.

Solution: Always use wait_for with a selector before clicking:

await navigate(sessionId, tabId, newUrl);
await wait_for(sessionId, tabId, selector: ".load-more-btn", ms: 5000);
await click(sessionId, tabId, ".load-more-btn");  // Now safe

Never assume content exists immediately after navigation.

3️⃣ Fill vs Type

• fill() — clears the field and types (uses React synthetic events)
• type_text() — types without clearing (for plain HTML)

Use fill() for modern frameworks. Use type_text() for simple inputs.

4️⃣ Snapshot + TOTP

Problem: TOTP codes change every 30 seconds. Can't snapshot mid-auth.

Solution: Log in manually, wait until you're past the TOTP challenge, *then* snapshot. The saved session includes all cookies — next restore won't need TOTP again.

5️⃣ Wait_For Ambiguity

wait_for can wait for a selector, a URL pattern, or a fixed delay. The response tells you what happened.

Read the metadata block. It shows _condition_type (selector/url/fixed_delay) and _condition_met (true/false).

6️⃣ Sessions Survive Daemon Restarts (v0.6.0+)

Problem: You restart the daemon expecting a clean slate, but existing Chrome windows are still attached.

Why: v0.6.0 uses TCP-based Chrome transport — Chrome runs independently of the daemon process. Sessions auto-reattach on startup.

Solution: This is intentional. Call list_sessions() to see surviving sessions. Call close_session(id) to clean up explicitly if you don't want them. See ADVANCED.md → "Session Persistence & Auto-Reattach".

Core Workflow: Form Filling with Error Recovery

Real-world example. Fill a React form, handle validation errors.

// 1. Open session
const sessionId = await open_session({ profile: "personal" });
const tabs = await list_tabs(sessionId);
const tabId = tabs[0].target_id;

// 2. Navigate to the form
await navigate(sessionId, tabId, "https://example.com/form");
await wait_for(sessionId, tabId, selector: ".submit-btn", ms: 5000);

// 3. Inspect the form structure
const content = await get_content(sessionId, tabId);
// Claude reads: email field, password, checkbox, submit button

// 4. Fill the form with error recovery
try {
  await fill(sessionId, tabId, "input[name='email']", "user@example.com");
  await fill(sessionId, tabId, "input[name='password']", "secret");
  await click(sessionId, tabId, "input[type='checkbox']");
  await click(sessionId, tabId, ".submit-btn");

  // Wait for success
  await wait_for(sessionId, tabId, selector: ".success-message", ms: 5000);

} catch (error) {
  // If validation error, read it and retry
  const errorMsg = await get_content(sessionId, tabId);
  // Claude parses "Email already taken" → uses different email → retry
}

// 5. Done
await screenshot(sessionId, tabId);
await close_session(sessionId);

Key patterns:

• Use fill() for React/Vue/Angular (synthetic events)
• Always wait_for before interacting with dynamic content
• Try-catch around the whole interaction block for recovery
• Screenshots as verification checkpoints

When to Use Pagerunner vs Other Tools

Deeper Dives

• I want to learn all the workflow patterns → PATTERNS.md
• I need to see all ~44 tools → REFERENCE.md
• I need to handle sensitive data (PII, credentials, sealed secrets) → SECURITY.md
• I want to call a site's API directly from my session (site adapters) → REFERENCE.md § Site Intelligence
• I want to inspect network / console traffic → REFERENCE.md § Network & Console
• I want to record the session as a video → RECORDING.md (director's guide) or REFERENCE.md § Video Recording for tool signatures
• Results look wrong? Hallucination? → HALLUCINATION_PREVENTION.md
• Show me full ICP workflows → EXAMPLES.md
• Multi-agent coordination, daemon setup, auto-recovery → ADVANCED.md
• Something broke? → DEBUGGING.md

Setup

1. Install Pagerunner

# macOS (Homebrew — easiest)
brew tap enreign/pagerunner
brew install pagerunner

# Or Cargo
cargo install pagerunner

2. Register as MCP server

claude mcp add pagerunner "$(which pagerunner)" mcp

For Claude Desktop, add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "pagerunner": {
      "command": "/path/to/pagerunner",
      "args": ["mcp"]
    }
  }
}

3. Configure Chrome profiles (optional)

pagerunner init

This reads your Chrome profiles and writes ~/.pagerunner/config.toml.

Note: Close any Chrome window before opening a Pagerunner session on that profile (Chrome locks directories).

Next Steps

Pick your ICP above, follow the quick start, then dive into the relevant doc:

• Solo Dev → PATTERNS.md → "Frontend dev loop"
• Power User → PATTERNS.md → "Authentication persistence"
• Security-Conscious → SECURITY.md
• Server-Side → ADVANCED.md → "Multi-agent patterns"

Happy automating!