Web Search 9routers Backup

changelog: Add ClawHub metadata, trigger-rich description, error handling, and red flags

Web Search 9routers Backup Skill

When to Use

User provides a URL to fetch/read, or the primary 9router-web-search skill fails. This wraps the internal 9Router backup web fetch service on localhost:20128.

Usage

Provide URL and optional parameters to fetch content via the internal 9routers backup web fetch service.

Parameters

• url: target URL to fetch
• model: model to use (default: exa)
• format: output format (html or text, default: html)
• max_characters: maximum characters to return (0 for unlimited, default: 0)
• authorization_token: bearer token (if not set, uses default from environment or config)

Example

bash

curl -X POST http://localhost:20128/v1/web/fetch \
 -H "Content-Type: application/json" \
 -H "Authorization: Bearer $NINEROUTER_KEY" \
 -d '{"model":"exa","url":"https://example.com","format":"html","max_characters":0}'

Implementation Notes

• This skill wraps the internal web fetch endpoint.
• Ensure the endpoint is accessible (localhost:20128) within the OpenClaw environment.
• The token may need to be refreshed periodically.

Error Handling

Scenario	Response
Connection refused on :20128	9Router backup service is down — report to user, suggest primary skill
HTTP 401/403	Token expired — ask user to refresh NINEROUTER_KEY
Empty content	Retry with format:text or report page as unreadable

Red Flags — STOP

• Never hardcode the bearer token in commands or output — use $NINEROUTER_KEY
• Never fetch URLs from untrusted user input without confirmation (SSRF risk)

Version

sha256:PLACEHOLDER