{"slug": "how-i-built-an-ai-journalist-discovery-engine-with-octoparse-mcp", "title": "How I Built an AI Journalist Discovery Engine with Octoparse MCP", "summary": "A developer built E_MediaScience, an earned media operating system that uses Octoparse MCP as a live intelligence agent to discover and match journalists to PR pitches. The system calls Octoparse MCP with intent-based parameters, returns structured journalist profiles, and feeds them into Claude for newsworthiness scoring and personalized outreach, all in under 60 seconds. A key innovation is the HungQueryResolver, which handles vague natural language queries through a three-strike escalation architecture.", "body_md": "Most people connect Octoparse MCP to their AI assistant and use it to extract a product list or pull some prices into a table.\n\nThat's fine. But I wanted to use it differently.\n\nI wanted Octoparse MCP to act as a live structured intelligence agent — something my AI system could call on demand, in real time, every time a user submits a story or a press angle. Not a batch job. Not a scheduled pipeline. A live tool call that returns clean, structured journalist profiles directly into an LLM scoring engine.\n\nThat's what I built with E_MediaScience.\n\nThe Problem\n\nPR intelligence tools like Cision and Muck Rack cost $10,000–$30,000 per year. They're inaccessible to most founders, startups, and SMEs — the exact people who most need earned media coverage to grow.\n\nThe alternative is hours of manual research: scanning publication mastheads, reading journalist bylines, guessing at beats and tone. Even then, the outreach is generic because there's no structured data behind it.\n\nThe core problem isn't finding journalists. It's vocabulary asymmetry.\n\nA founder knows their product. They don't know how a journalist at TechCrunch would classify it, what beat editor covers their category, or which publications have recently covered adjacent topics. Traditional search tools enforce a tight validation loop — keep refining your query until you find something — and most users give up before they get there.\n\nWhat E_MediaScience Does\n\nE_MediaScience is a multi-tenant earned media operating system. A user submits a story, launch, or campaign brief. The system:\n\nCalls Octoparse MCP with intent-based parameters — not a URL, but a topic and geographic target\n\nOctoparse selects the appropriate journalist discovery template from its 600+ library and executes geo-routed extraction\n\nReturns clean, structured journalist profiles: name, outlet, beat, article history, tone markers, contact data\n\nFeeds that payload directly into Claude for AI newsworthiness scoring and journalist matching\n\nGenerates personalised outreach referencing each journalist's actual recent work\n\nTracks replies, open rates, and campaign strike rate\n\nThe entire flow takes under 60 seconds from submission to matched journalist list.\n\nWhy Octoparse MCP Changes Everything\n\nBefore MCP, my options were:\n\nBuild and maintain custom scrapers per publication (brittle, expensive, breaks constantly)\n\nUse a static journalist database (stale, expensive, no real-time beat tracking)\n\nAsk an LLM to find journalists (hallucinated profiles, made-up contact details)\n\nOctoparse MCP eliminates all three problems in a single tool call.\n\ntext\n\nUser submits: \"I've launched an AI video clipping tool for live-sellers\"\n\nEMS calls Octoparse MCP:\n\n→ Template: journalist-discovery-tech-ecommerce\n\n→ Parameters: { topic: \"AI video tools, live commerce, creator economy\", regions: [\"UK\", \"US\"] }\n\nOctoparse returns:\n\n→ 12 journalist profiles, structured JSON\n\n→ Beat: \"Commerce technology, live shopping, creator tools\"\n\n→ Recent articles, outlet, contact data — all clean, no parsing\n\nClaude scores:\n\n→ Newsworthiness: 74/100\n\n→ Top match: [Journalist at The Information, beat: AI/Creator Economy]\n\n→ Personalised pitch: references journalist's last 3 articles\n\nNo HTML. No CSS selectors. No fragile extraction logic. The structured payload goes straight into the LLM.\n\nThe HungQueryResolver — V1.1 Innovation\n\nThe most technically novel part of E_MediaScience is the HungQueryResolver — built specifically around what Octoparse MCP can do when queries fail.\n\nThe problem: clients describe their PR targets in natural language that doesn't map cleanly to journalist taxonomies. \"Find people who write about the neat tech stuff I make\" is a real query. Traditional systems force clarification loops until the user gives up.\n\nThe HungQueryResolver uses a Three-Strike Escalation architecture:\n\nTurn 1 — Direct Match\n\nOctoparse MCP called with the raw query. High-confidence matches are returned immediately.\n\nTurn 2 — Drift Validation\n\nIf confidence falls below threshold, the user is prompted once for clarification. The system measures whether the new query actually adds new information — or just rephrases the same intent.\n\nTurn 3 — Async Escalation\n\nIf the user is circling the same concept in different words, the system stops asking. A background worker fires a broadened Octoparse MCP call with expanded terminology, adjacent industry classifications, and alternate journalist taxonomies — silently, while the UI holds.\n\nInstead of a dead end, the user gets a scored set of alternative matches with a transparent quality rating explaining why each result was surfaced.\n\nThis turns a search failure into a consulting asset.\n\nThe Multi-Tool MCP Stack\n\nE_MediaScience was built in Cursor IDE using Claude Sonnet and Opus. The full MCP stack:\n\nOctoparse MCP — Structured journalist extraction (primary data source)\n\nSupabase MCP — Schema management, RLS policies, Edge Function deployment\n\nGitHub MCP — Automated commits across the GlafyCo org\n\nGlobalProxyManager — Custom geo-routing layer for multi-region journalist discovery across 100+ geographic IPs\n\nThe combination of Octoparse (extraction) + Claude (reasoning) + Supabase (persistence) creates a closed-loop intelligence system where every journalist match is grounded in real, live web data.\n\nProduction Philosophy — Pipeline Not Repository\n\nOne design decision worth sharing: Octoparse MCP is never used as a data warehouse.\n\nEvery extraction is immediately scored, matched, and actioned. Data follows a strict TTL policy:\n\nDays 1–30: Hot storage — full access, edit, download\n\nDays 31–60: Cold storage — read-only, raw source stripped\n\nDay 61: Hard delete\n\nThis keeps infrastructure lean and reinforces the product positioning: E_MediaScience is a processing engine, not a data repository. Users ingest, score, pitch, and clear the decks.\n\nPricing Model\n\nE_MediaScience uses a Core + Engines modular pricing architecture:\n\nCore Platform — $69/month (dashboard, 2 seats, campaign management)\n\nSignal Engine bolt-on (EMS) — from $29/month\n\nProduction Engine bolt-on (Clipositing video engine) — from $29/month\n\nAgency tiers — from $799/month with HighLevel CRM integration\n\nNo credits. No per-minute charges. No \"credit hostage-taking.\" Flat session-based pricing that scales by tier, not by usage clock.\n\nRepo\n\nEverything is open and committed:\n\nGitHub: github.com/GlafyCo/E_MediaScience\n\nThe architecture docs, sprint plans, and HungQueryResolver spec are all in docs/strategy/. The multi-tenant core, AI scoring engine, and Supabase migrations are all there.\n\nBuilt with Octoparse MCP + Cursor + Claude for the Octoparse MCP Challenge 2026.\n\nIan Taylor — Founder, GlafyCo | Wales, UK\n\nBuilding E_MediaScience, Clipositing, and the GlafyCo AI platform stack | X: [@ianbuildsagents](https://dev.to/ianbuildsagents)", "url": "https://wpnews.pro/news/how-i-built-an-ai-journalist-discovery-engine-with-octoparse-mcp", "canonical_source": "https://dev.to/ianbuildsagents/how-i-built-an-ai-journalist-discovery-engine-with-octoparse-mcp-3im", "published_at": "2026-06-13 07:16:14+00:00", "updated_at": "2026-06-13 07:47:56.553026+00:00", "lang": "en", "topics": ["ai-tools", "artificial-intelligence", "large-language-models", "developer-tools"], "entities": ["Octoparse MCP", "E_MediaScience", "Claude", "Cision", "Muck Rack", "TechCrunch", "The Information"], "alternates": {"html": "https://wpnews.pro/news/how-i-built-an-ai-journalist-discovery-engine-with-octoparse-mcp", "markdown": "https://wpnews.pro/news/how-i-built-an-ai-journalist-discovery-engine-with-octoparse-mcp.md", "text": "https://wpnews.pro/news/how-i-built-an-ai-journalist-discovery-engine-with-octoparse-mcp.txt", "jsonld": "https://wpnews.pro/news/how-i-built-an-ai-journalist-discovery-engine-with-octoparse-mcp.jsonld"}}