Skip to content
Back to skills

Haunt Web Extraction

ASecurity

Use Haunt MCP to extract structured JSON, clean Markdown, article fields, metadata, or usage from permitted public web pages.

  • 3 stars
  • 0 votes
  • 0 copies
  • 5 views
  • Added September 12, 2026
ai-agentsgogitapi

Works with

  • claude code
  • cursor
  • api
  • mcp

Security analysis

A100/100

Scanned September 12, 2026

npx -y skills add thedixitjain/the-mega-skill-library --skill haunt-web-extraction --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Haunt Web Extraction?

Add the live security badge to your README. It updates with every re-scan.

Security grade badge for Haunt Web Extraction
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/thedixitjain-haunt-web-extraction/badge)](https://www.skillsdirectory.com/skills/thedixitjain-haunt-web-extraction)

More formats (shields.io, HTML) on the badges page. Keep it an A: scan every change in CI with Pro.

Download with Pro
SKILL.md
---
name: haunt-web-extraction
description: "Use Haunt MCP to extract structured JSON, clean Markdown, article fields, metadata, or usage from permitted public web pages."
category: mcp-and-integrations
source_repo: hashgraph-online/awesome-codex-plugins
source_path: "plugins/Darko893/haunt-mcp-server/skills/haunt-web-extraction/SKILL.md"
source_url: https://github.com/hashgraph-online/awesome-codex-plugins/blob/HEAD/plugins/Darko893/haunt-mcp-server/skills/haunt-web-extraction/SKILL.md
---


# Haunt Web Extraction

Use this skill when the user needs data from a permitted public web page inside Claude Code, Cursor, Codex, OpenCode, or another MCP-aware coding agent.

## First action

If the Haunt MCP server is installed but no API key is configured yet, call `try_demo_extract` first. It proves the server is wired correctly and returns docs, signup, pricing, MCP info, and free-tier links without using credits.

## Live extraction

For live pages, require `HAUNT_API_KEY` in the MCP server environment. Use:

- `extract_url` or `extract` for structured JSON from a permitted public URL.
- `extract_markdown` for readable page text for RAG, notes, docs ingestion, or `.md` files.
- `extract_article` for public article pages.
- `extract_metadata` for titles, descriptions, Open Graph, Twitter Card, and canonical URL metadata.
- `get_usage` after live calls to check plan and remaining credits.

## Boundaries

Haunt is for permitted public pages and supported rendered pages, including many that sit behind Cloudflare or a bot wall. It does not promise CAPTCHA solving, login-wall access, paywall access, or restricted-page access. If a page is blocked, login-required, CAPTCHA-gated, paywalled, restricted, or too thin to verify, treat Haunt's clear failure signal as the correct result. Do not fabricate extracted data.

## Good prompt pattern

"Use Haunt to extract the product name, price, availability, and review count from this permitted public URL: <url>. Return JSON and tell me if the page was blocked or unverifiable."

---

**Source:** [`hashgraph-online/awesome-codex-plugins`](https://github.com/hashgraph-online/awesome-codex-plugins) → `plugins/Darko893/haunt-mcp-server/skills/haunt-web-extraction/SKILL.md`

Attribution

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments

Loading comments…

Newsletter

Get the weekly list of safe new skills, and the ones we pulled.