actionbook-scraper
Generate and verify web scraper scripts using Actionbook's verified selectors. Auto-validates generated scripts and fixes errors.
Profile →npx skills add actionbook/actionbook --skill actionbook-scraper --agent claude-code
Same command for any agent — swap --agent for codex, cursor, copilot.
Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.
What it does
Generates a scraper script based on Actionbook selectors and then verifies it by running the script and checking two aspects: (1) the script runs without errors or timeouts, and (2) the extracted data matches the expected content (not empty, not placeholder or UI text, with correctly mapped fields). If issues arise, it logs Actionbook data issues and retries up to three times, aiming to produce a verified script and a data preview.
How it works
- The workflow requires two-part verification: Part 1 ensures the script runs without errors/timeouts; Part 2 ensures the extracted data contains the correct fields and values.
- It instructs generation via Actionbook selectors, then execution to confirm behavior. If failures occur, it logs to a local issue file and retries (up to 3x).
- It supports two output formats: a default agent-browser bash command sequence and a standalone Playwright JavaScript file when using the --standalone option. It also enforces closing the browser after execution.
- It emphasizes including scroll handling for lazy-loaded content and expand/collapse logic for card-based layouts, and uses Actionbook MCP tools as the first source for selectors.
When to use it
- Use when you need a scraper script generated from verified selectors and you require automated verification that the script runs and extracts valid data.
- Use the default agent-browser mode for quick, command-based scraping flows; use --standalone to obtain a Playwright script.
What it can touch
- The skill references the following tools and outputs but does not expose explicit touch points beyond the generated scripts and the verification steps:
- agent-browser commands (open, wait, scroll, get text, close)
- Playwright code when using --standalone
- Local log file .actionbook-issues.log for data issues
Caveats
- Requires both checks to pass: script runs and data content correctness.
- If data validation fails, the recommended action is to fix selectors, wait logic, or extraction, then retry up to 3x.
- The approach depends on Actionbook-provided selectors and may log issues if selectors are outdated or incorrect.
# Actionbook Scraper Skill ## ⚠️ CRITICAL: Two-Part Verification **Every generated script MUST pass BOTH checks:** | Check | What to Verify | Failure Example | |-------|----------------|-----------------| | **Part 1: Script Runs** | No errors, no timeouts | `Selector not found` | | **Part 2: Data Correct** | Content matches expected | Extracted "Click to expand" instead of name | ``` ┌─────────────────────────────────────────────────────┐ │ 1. Generate Script │ │ ↓ │ │ 2. Execute Script │ │ ↓ │ │ 3. Check Part 1: Script runs without errors? │ │ ↓ │ │ 4. Check Part 2: Data content is correct? │ │ - Not empty │ │ - Not placeholder text ("Loading...") │ │ - Not UI text ("Click to expand") │ │ - Fields mapped correctly │ │ ↓ │ │ ┌───┴───┐ │ │ BOTH Pass Either Fails │ │ │ │ │ │ │ ↓ │ │ │ Is it Actionbook data issue? │ │ │ │ │ │ │ ┌───┴───┐ │ │ │ Yes No │ │ │ │ │ │ │ │ ↓ ↓ │ │ │ Log to Fix script │ │ │ .actionbook-issues.log │ │ │ │ │ │ │ │ └───┬───┘ │ │ │ ↓ │ │ │ Retry (max 3x) │ │ ↓ │ │ Output Script │ └─────────────────────────────────────────────────────┘ ``` ## Default Output Format ``` /actionbook-scraper:generate <url> ``` **DEFAULT = agent-browser script (bash commands)** ```bash
- ⚠️ CRITICAL: Two-Part Verification
- Default Output Format
- With --standalone Flag
- Verification Requirements
- Two-Part Verification
- Part 1: Script Execution Check
- Part 2: Data Content Check (CRITICAL)
- For agent-browser Scripts
- For Playwright Scripts (--standalone)
- Architecture Overview
- Tool Priority
- Workflow Rules
- CRITICAL: Generate → Verify → Fix
- Verification Process
agent-browser open "https://example.com" agent-browser scroll down 2000 agent-browser get text ".selector" agent-browser close agent-browser open "https://example.com/page" agent-browser wait --load networkidle agent-browser get text ".item-container" npm install playwright node scraper.js agent-browser open "https://firstround.com/companies"
What does the actionbook-scraper skill do?
Generate and verify web scraper scripts using Actionbook's verified selectors. Auto-validates generated scripts and fixes errors.
How do I install it?
Run `npx skills add actionbook/actionbook --skill actionbook-scraper --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.
Where does this skill come from?
From actionbook/actionbook, a repository with 1,584 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.
Is a popular skill a good skill?
Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.