ContentOps — Open Recipe Architecture

Closed-loop recipe performance intelligence → automated content generation → distribution

ContentOps Platform DATA SOURCES Social Media / Web TikTok · Instagram · Reddit Pinterest · YouTube · Blogs Each post has engagement data AI Search Results ChatGPT · Perplexity Google AI Overviews Citation tracking Blog Post / Article (DC Agent Entry Point) Recipe extracted from blog post content INGESTION & CRAWLING LAYER Nimble 🔍 Web Scraper Recipe text · views · saves Senso 📋 Citation Tracker ChatGPT · Perplexity · AIO raw posts + engagement data AI citation data POST → RECIPE MATCHING 🤝 Post → Recipe Matching (Semantic Search) For every incoming social post / article → extract content → Senso KB semantic search → find closest recipe match Senso auto-embeds all recipes. Query: "find recipe most similar to this [post content]" → returns matched recipe ID + similarity score scraped post content blog post → extract → match semantic search THE OPEN RECIPE STANDARD (recipe.md) recipe.md — Open Source Recipe Standard Any agent can read this format. Accumulates performance intelligence over time. General Stats engagement rate AI citation count saves / shares Gen Instructions prerequisites content blocks format rules Linked Posts output history (1:M) platform + brand performance per post matched recipe + perf data embed + store analytics ClickHouse 🏠 Analytics DB Recipe embeddings cosineDistance search Performance trends Engagement analytics ANN indexes Senso KB 📋 Knowledge Graph Auto-embedded recipes Semantic search engine Brand context store cited.md publishing agent-ready API queryable via API semantic search results DC AGENT — CONTENT GENERATION & DISTRIBUTION Open Recipe API /search · /trending · /similar DC Agent 1. Blog URL → extract recipe 2. Query API for best format DeepContent Gen Reddit post · Carousel Video script · AI snippet Distribution cited.md · Social Reddit · Cross-platform query generate publish 🔄 Closed Loop: published content feeds back as data for the next scrape → continuous improvement Datadog Lapdog 🐶 Observability Traces every step: → Nimble scrape → Senso citation → Post→Recipe match → Recipe embed/vector → ClickHouse query → DC generation → cited.md publish ────────────── Live demo in browser Sponsor Tool (Nimble · Senso · ClickHouse) DC Agent / API Open Recipe Standard Datadog Lapdog 🔄 Closed Loop

Sponsor Tools — 4 Used

  • • Nimble 🔍 — Scrapes recipe performance data from social platforms + AI search
  • • Senso 📋 — Citation tracking, KB with auto-embedding, semantic post→recipe matching, cited.md publish
  • • ClickHouse 🏠 — Recipe embeddings, vector similarity (cosineDistance), performance analytics
  • • Datadog Lapdog 🐶 — Full observability, traces every span across the entire pipeline

recipe.md Open Standard

  • • General Stats — Engagement rates, AI search citations, saves, shares — accumulates signal over time
  • • Gen Instructions — Prerequisites, content blocks, rules, format — the static recipe definition
  • • Linked Posts — Output history: every post this recipe generated, platform, brand, performance
  • • Recipe : Post = 1 : Many — each post maps to exactly one recipe via semantic matching

DC Agent Demo Flow

  • • Hand it a blog post URL → extract recipe content
  • • Post → Recipe Matching: Senso KB semantic search finds closest recipe
  • • Query Open Recipe API for that recipe's performance data
  • • DeepContent generates right format (Reddit, carousel, video script)
  • • Publish to cited.md → closes the loop → feeds back for next scrape

ContentOps · Agentic Engineering Hack · May 23, 2026