citedy-content-ingestion

name	citedy-content-ingestion
title	Content Ingestion
description	Turn any URL into structured content — YouTube videos (via Gemini Video API), web articles, PDFs, and audio files. Extract transcripts, summaries, and metadata for use in any LLM pipeline. Powered by Citedy.
version	1.0.0
author	Citedy
tags	["content-ingestion","youtube","transcription","pdf","audio","web-scraping","data-extraction"]
metadata	{"openclaw":{"requires":{"env":"[Truncated]"},"primaryEnv":"CITEDY_API_KEY"},"compatible_with":"citedy-seo-agent@3.2.0"}
privacy_policy_url	https://www.citedy.com/privacy
security_notes	API keys (prefixed citedy_agent_) authenticate against Citedy API endpoints only. All traffic is TLS-encrypted. Keys can be revoked from dashboard.

Content Ingestion — Skill Instructions

Connection: REST API over HTTPS Base URL: https://www.citedy.com Auth: Authorization: Bearer $CITEDY_API_KEY

---

Overview

Turn any URL into structured content your agent can use. Pass a link — the skill extracts the full text, transcript, metadata, and summary — and returns it as clean structured data ready for your LLM pipeline.

Supported content types:

• YouTube videos — full transcription via Gemini Video API (not just captions)
• Web articles — clean article text with metadata
• PDF documents — text extraction from public PDF URLs
• Audio files — transcription from MP3/WAV/M4A files

Differentiator: YouTube ingestion uses the Gemini Video API for deep video understanding — it goes beyond auto-generated captions, capturing speaker intent, visual context, and structure.

Use this skill as a standalone input node for any LLM pipeline. Feed the output directly into summarization, Q&A, article generation, or knowledge base indexing.

---

When to Use

Use this skill when the user:

• Asks to extract, transcribe, or summarize a URL
• Shares a YouTube video and wants the content analyzed or repurposed
• Shares a PDF link and wants the text extracted
• Wants to ingest audio content for transcription
• Is building a pipeline that needs to pull content from the web

---

Instructions

Setup (run once)

If you don't have a saved API key for Citedy, run this flow:

1. Register

Preferred: run the included registration script:

node scripts/register.mjs [agent_name]

The script calls the registration API and prints the approval URL. If agent_name is omitted, it defaults to agent-<hostname>.

Alternative: call the API directly:

POST https://www.citedy.com/api/agent/register
Content-Type: application/json

{"agent_name": "<your_agent_name>"}

Either way, you'll get back:

{
  "pending_id": "..."