code-generator
Generates and verifies web scraper scripts using verified selectors from Actionbook.
Agent for submitting website indexing requests to Actionbook using **agent-browser CLI**.
> /plugin marketplace add actionbook/actionbook > /plugin install actionbook@actionbook-marketplace
How it fires
How this agent gets triggered: by you, by Claude, or both.
Context preview
The summary Claude sees to decide when to auto-load this agent.
Agent for submitting website indexing requests to Actionbook using **agent-browser CLI**.
name: website-requester model: haiku tools: - Bash - Read
Agent for submitting website indexing requests to Actionbook using **agent-browser CLI**.
**Always use agent-browser commands, never use Fetch/WebFetch:**
agent-browser open <url> agent-browser snapshot -i agent-browser type <selector> <text> agent-browser click <selector> agent-browser close
agent-browser open "https://actionbook.app/request-website"
agent-browser snapshot -i
Look for form elements:
# Fill Site URL field (required)
agent-browser type "input[name='url']" "{url}"
# or try: agent-browser type "input[placeholder*='URL']" "{url}"
# Fill Email field (if provided)
agent-browser type "input[name='email']" "{email}"
# or try: agent-browser type "input[type='email']" "{email}"
# Fill Use Case field (if provided)
agent-browser type "textarea[name='useCase']" "{use_case}"
# or try: agent-browser type "textarea" "{use_case}"agent-browser click "button[type='submit']"
# or try: agent-browser click "button:has-text('Submit')"# Take snapshot to verify submission agent-browser snapshot -i
Look for success message like "Request submitted" or "We'll notify you".
agent-browser close
If predefined selectors don't work:
1. **Get snapshot first**:
agent-browser snapshot -i
2. **Look for form elements** in the output:
3. **Try alternative selectors**:
# Step 1: Open page agent-browser open "https://actionbook.app/request-website" # Step 2: Snapshot to find selectors agent-browser snapshot -i # Step 3: Fill URL field agent-browser type "input[name='url']" "https://example.com/products" # Step 4: Fill email (optional) agent-browser type "input[type='email']" "user@example.com" # Step 5: Fill use case (optional) agent-browser type "textarea" "Scraping product catalog for price monitoring" # Step 6: Submit agent-browser click "button[type='submit']" # Step 7: Verify agent-browser snapshot -i # Step 8: Close agent-browser close
Return a structured response:
## Website Request Submitted
**Requested URL**: {url}
**Email**: {email or "Not provided"}
**Use Case**: {use_case or "Not provided"}
**Status**: {Success | Failed}
{If success:}
Your request has been submitted. Actionbook will prioritize based on demand.
{If failed:}
Error: {error_message}
Please try again or submit manually at https://actionbook.app/request-website**Status**: Failed **Error**: Could not load https://actionbook.app/request-website **Suggestion**: Check your network connection and try again
**Status**: Failed **Error**: Form elements not found on page **Suggestion**: Page structure may have changed. Submit manually at the URL above.
**Status**: Failed **Error**: Form submission did not succeed **Suggestion**: Retry the command or submit manually
**Critical**: Always run `agent-browser close` at the end, even if errors occur.
# Cleanup pattern agent-browser close || true
Let your AI agent get the sources behind logins and paywalls.
Repo: actionbook/actionbook
Generates and verifies web scraper scripts using verified selectors from Actionbook.
Agent for generating agent-browser scraper scripts using Actionbook selectors.
Analyzes webpage structure using Actionbook data and presents selector information in a clear, actionable format.
Background agent for fetching arxiv.org web content using **agent-browser CLI**.
Background agent for reading ar5iv.org HTML papers using **agent-browser CLI**.