markdown-browser

Aggregate and sanitize web_fetch results by enforcing policy and redaction rules.

Updated Mar 4, 2026
One-click install
npx skills add https://github.com/velamints2/clawbot-lab --skill markdown-browser
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: markdown-browser
Source: https://github.com/velamints2/clawbot-lab/tree/main/skills/markdown-browser
Command: npx skills add https://github.com/velamints2/clawbot-lab --skill markdown-browser

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires turndown.

What problem does it solve?

This wrapper orchestrates OpenClaw web_fetch results to ensure policy compliance, privacy redaction, and consistent output formatting, so agents can reason safely on fetched content without leaking data.

Core Features & Use Cases

  • Policy enforcement: automatically interprets Content-Signal headers to decide whether to allow, block, or flag inputs.
  • Privacy redaction: redacts sensitive URL components to protect secrets before they reach downstream logic.
  • Content normalization: preserves Markdown when available, or converts HTML to Markdown via a fallback for stable downstream rendering.
  • Output standardization: emits a single, structured schema with content, format, policy_action, source_url, status_code, and token estimates for reproducible integration.

Quick Start

Run the wrapper on a web_fetch payload to produce a normalized, redact-safe output ready for downstream agents.

Frequently Asked Questions about markdown-browser

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I sanitize web_fetch results to enforce content policy and redact sensitive URL parts?▼

To sanitize web_fetch results, you can use a wrapper that interprets Content-Signal headers to enforce policy and redacts sensitive URL components. It outputs structured, redact-safe markdown or HTML-fallback content for downstream agents.

Why does my web_fetch output need markdown normalization before downstream processing?▼

Markdown normalization ensures consistent formatting for downstream agent reasoning by preserving markdown when available or converting HTML via a fallback. This standardization prevents rendering instability and provides reproducible structured outputs.

Can I use turndown to convert HTML to Markdown for web_fetch payloads?▼

Yes, the process uses turndown as a fallback to convert HTML to Markdown for fetched web content. This preserves formatting stability when native markdown is unavailable in the web_fetch payload.

Does this approach provide a structured output schema for fetched web content?▼

Yes, the wrapper emits a single structured output schema containing content, format, policy_action, source_url, status_code, and token estimates. This standardization ensures reproducible integration for downstream processing.

What is the best way to prevent data leakage when agents process fetched web content?▼

Preventing data leakage requires orchestrating web_fetch results with privacy redaction and policy enforcement. The wrapper redacts sensitive URL components and flags or blocks inputs based on Content-Signal policy before content reaches agents.

When should I not use a wrapper for web_fetch content normalization?▼

You should not use a wrapper if your fetched content requires no policy enforcement, privacy redaction, or schema standardization. It is specifically designed for sanitizing and structuring payloads to ensure safe downstream agent reasoning.