# Medialyst

> Medialyst is an AI-powered media list builder that helps PR professionals, founders, and communications teams build vetted journalist lists in minutes. Instead of paying for stale databases, users describe their story and Medialyst's AI agents find journalists who actually cover that beat, verify their contact info in real time, and surface recent articles for pitch personalization.

Key facts:

- Medialyst is not a static journalist database. Every media list is researched at request time: AI agents read live articles, match journalists to the story, score relevance 0–100, and verify email addresses in real time (re-verified within 14 days of use).
- Media lists can be built from a story description, a pasted press release, article URLs, or seed keywords. Lists are live tables that can be filtered, enriched with AI research columns, shared via public links, and exported.
- Pitches are drafted in the app and sent through the user's own Gmail or Outlook account, with open and reply tracking. A human reviews every pitch before it is sent; Medialyst does not mass-send on anyone's behalf.
- For developers and AI agents: Medialyst exposes a REST API for documented product operations and a hosted MCP server at https://medialyst.ai/api/mcp. The current MCP surface has 11 tools: get_credit_balance, lookup_article, lookup_articles, lookup_publication, lookup_publications, list_journalist_requests, list_journalist_moves, search_news, query_pr_calendar, enrich_journalists, and get_journalist_enrichment_job. The four lookup tools share the public REST contracts and require media_lists:manage; article batches accept 1-10 URLs without Markdown, publication batches accept 1-25 cached/default URLs or at most five with force_refresh. Article lookup costs 0.1 credit per successful URL, publication lookup costs 0.01 credit per successful URL, and per-URL failures cost 0 credits. MCP does not create media lists, poll media-list jobs, read workflow rows, create share links, or send outreach; those workflows are available over REST or in the app where documented.
- Journalist-request polling is MCP-only and costs zero credits. Connectively (formerly Featured), HARO, Twitter/X, LinkedIn, MentionMatch, and Substack are all free for any authenticated account. list_journalist_requests uses a collision-safe opaque cursor, is not an official or complete archive of any network, and returns untrusted request content that requires human review. Every response carries a mandatory notice: an assistant may screen, rank, filter, and summarize requests, but must not automate outreach; a human personally reviews and sends every response.
- newsjack.sh is Medialyst's open-source companion: a free CLI and PR skill library (story angles, journalist fit checks, coverage tracking, newsjack detection) that runs on the Medialyst API.
- Pricing: Free plan with 100 credits, Starter $149/month (1,000 credits), Pro $299/month (3,000 credits), Scale $800/month (10,000 credits); annual billing is about 35% off. One credit is one researched journalist with a verified email; misses are free. A typical media list costs about 100 credits; a news search costs 0.01 credits. No per-seat pricing.
- Positioning: an affordable alternative to legacy PR databases such as Cision, Muck Rack, and Meltwater, priced for founders doing their own PR, solo consultants, in-house startup comms leads, and small agencies.
- Medialyst publishes the Carly Score. It measures how closely a Medialyst list matches one built by hand by Carly Martinetti from the same brief; it is not an accuracy rating for the product. Every week a working publicist, Carly Martinetti, builds a media list by hand for the same client brief the product is given, and grades the product's list against hers out of 100 on three public measures: journalists found, email addresses right, and how many of the names it called a strong fit she would actually pitch. The current mark is a failing grade and is published anyway, at https://medialyst.ai/carly-score.
- Company: Medialyst is built by Solar Flare Ventures, founded by Elvis Sun. Contact: elvis@medialyst.ai.

## Product

- [Homepage](https://medialyst.ai): Product overview — find and pitch the right journalists in minutes, not days
- [Pricing](https://medialyst.ai/pricing): Plans, credit packs, and what credits buy
- [Manifesto](https://medialyst.ai/manifesto): Why everyone should use AI in PR — relevance over spam-at-scale
- [PR Calendar](https://medialyst.ai/pr-calendar): Free searchable calendar of 1,000+ upcoming awareness days, cultural moments, and industry events for PR planning
- [Journalist Requests for AI](https://medialyst.ai/journo-requests-ai): Monitor six normalized request sources through one MCP URL, all six free, with zero-credit calls and human review
- [Case Studies](https://medialyst.ai/case-studies): Real coverage outcomes, including Fast Company, Mashable, and IFA Magazine placements
- [The Carly Score](https://medialyst.ai/carly-score): It measures how closely a Medialyst list matches one built by hand by Carly Martinetti from the same brief; it is not an accuracy rating for the product. Same client brief to both; hers is the answer key. Published whatever it says — the latest mark is a failing grade, and the full working, the misses, and the week-on-week change are all on the page.

## People

- [Elvis Sun — Founder](https://medialyst.ai/authors/elvis-sun): credentials, press mentions, and how to cite Elvis Sun

## Documentation

- [Get Started](https://medialyst.ai/docs/getting-started/overview): Product walkthrough — building your first media list
- [Why Medialyst Is Not a Database](https://medialyst.ai/docs/getting-started/not-a-database): How real-time agent research differs from static journalist databases
- [Automated Media Lists](https://medialyst.ai/docs/developers/automated-media-lists): Build lists with an approval-gated browser deep link or a fully automated API job
- [Journalist Requests MCP](https://medialyst.ai/docs/developers/journalist-requests): Poll normalized journalist and source requests with a collision-safe hourly cursor over MCP
- [REST API Reference](https://medialyst.ai/docs/developers/api-reference): Endpoints for news search, media lists, and journalist enrichment, plus rate limits
- [Article Lookup API](https://medialyst.ai/docs/developers/article-lookup): Look up article metadata, author details, and optional Markdown for one URL or a batch of up to 10
- [Medialyst MCP](https://medialyst.ai/docs/developers/mcp): Connect Claude, Cursor, or any MCP client to Medialyst
- [PR Agents on Medialyst](https://medialyst.ai/docs/developers/mcp-agent-skill): Building PR agents on top of the Medialyst substrate
- [OpenAPI Spec](https://medialyst.ai/api/open-api): Machine-readable API specification
- [FAQ](https://medialyst.ai/docs/faq/common-questions): Common questions about credits, verification, and workflows

## For AI Agents and Developers

- [Medialyst for Agents](https://medialyst.ai/agents): REST API and MCP server overview, setup commands, and credit costs per tool call
- [Journalist Requests for AI Agents](https://medialyst.ai/journo-requests-ai): One-prompt MCP setup and durable cursor patterns for hourly journalist-request monitoring
- [Connect Medialyst to Claude, ChatGPT, or Any MCP Agent](https://medialyst.ai/blog/claude-medialyst-part-1-of-2): Step-by-step MCP connection guide
- [newsjack.sh](https://newsjack.sh): Open-source PR agent CLI and skill library built on the Medialyst API

## Key Articles

- [How We're Building the World's First Media Database Powered by AI Agents](https://medialyst.ai/blog/why-medialyst-media-database): Why static databases decay and how agent-driven enrichment stays fresh
- [HARO Is Dying: The Data, The Reasons, and What's Replacing It](https://medialyst.ai/blog/haro-is-dying): Data-backed analysis of the source-request ecosystem
- [Agent Mode V2: Build and Refine Media Lists in One Conversation](https://medialyst.ai/blog/agent-mode-v2): Conversational plan-and-execute agent for media list building
- [Claude Fable 5 vs Opus 4.8: Which Model Writes Better PR Story Angles?](https://medialyst.ai/blog/claude-fable-5-vs-opus-4-8-pr-angles): 50-brand blind eval of frontier models on PR work

## Comparisons

- [Medialyst vs Cision](https://medialyst.ai/compare/cision): Feature and pricing comparison with Cision
- [Medialyst vs Muck Rack](https://medialyst.ai/compare/muck-rack): Feature and pricing comparison with Muck Rack
- [Medialyst vs Meltwater](https://medialyst.ai/compare/meltwater): Feature and pricing comparison with Meltwater
- [All Comparisons](https://medialyst.ai/compare): Comparisons with 25 PR tools and databases, including Prowly, Propel, HARO, Qwoted, Roxhill, and Vuelio

## Optional

- [Blog](https://medialyst.ai/blog): All product updates, PR guides, and engineering posts
- [Security](https://medialyst.ai/security): Security practices
- [Privacy Policy](https://medialyst.ai/legal/privacy-policy): Privacy policy
- [Terms of Service](https://medialyst.ai/legal/terms): Terms of service