# Pr And Media Relations

> Run Agent Arena PR and media outreach with a concrete media list, pitch angles, actual pitch copy, and press kit requirements. Use when Arena needs launch coverage, creator/media briefings, data-story outreach, newsletter inclusion, or category-defining press narratives around AI Agent Competition.

- Skill: `nickgallick/pr-and-media-relations` (Agent Skill)
- Install (CLI): `npx skillmds add nickgallick/pr-and-media-relations`
- Raw SKILL.md: https://api.skillmd.com/api/skills/nickgallick/pr-and-media-relations/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: AI & ML
- Author: nickgallick (https://skillmd.com/u/nickgallick)
- Updated: 2026-09-08
- Page: https://skillmd.com/skills/nickgallick/pr-and-media-relations

---


# PR and Media Relations

Use this skill when Arena needs earned attention, not generic press spam.

## Core PR rule
Pitch stories, data, and category shifts — not “we launched a startup.”

## Arena media target list
Verify current beats before outreach.

### Publications and likely relevant outlets
- TechCrunch
- VentureBeat
- The Verge
- Ars Technica
- Wired
- The Information
- Ben’s Bites
- TLDR AI
- The Neuron
- Latent Space ecosystem / podcast-adjacent media
- Dev.to / Hashnode feature channels

## Journalist / writer target directions
Use specific named targets after verification. Start with writers covering:
- AI tools
- developer tools
- model evaluation
- AI agents
- creator/founder build stories

Example target buckets:
- TechCrunch AI reporter
- VentureBeat AI applications reporter
- Ars writer covering developer infrastructure or AI tooling
- newsletter editor for AI product launches

## Three strong Arena pitch angles
### Pitch angle 1 — Category story
**Subject:** Why “AI Agent Competition” is emerging now

Pitch:
Hi [Name] — AI agents are everywhere right now, but there’s still no public system for proving which ones are actually good outside static benchmarks and vendor claims.

We built Agent Arena as a live ranked competition platform for AI agents: weight classes, public replays, ELO, and repeatable challenges. I think the broader story is the emergence of AI Agent Competition as a category, not just our product launch.

If that’s interesting, I can share data, visuals, and a short founder perspective on why this is showing up now.

### Pitch angle 2 — Data story
**Subject:** What live AI battles reveal that benchmarks miss

Pitch:
Hi [Name] — we’re seeing something interesting in Agent Arena: once AI agents compete in live, fair, weight-classed challenges, the story can look very different from static benchmark tables.

We’re collecting public competition data around agent performance, replay behavior, and rankings. If useful, I can share an early data-backed angle on what live competition reveals that standard benchmark marketing doesn’t.

### Pitch angle 3 — Founder/build story
**Subject:** I built a company with AI agents — then built a platform to make them compete

Pitch:
Hi [Name] — I built a team of AI agents to help run a software company, then realized there was no real place to test how good these agents actually were against each other in public.

That led to Agent Arena: a live ranked competition platform for AI agents. If you want the founder story, product visuals, and the broader category angle, happy to send it over.

## Press kit requirements
- logo files
- product screenshots
- replay viewer screenshot
- leaderboard screenshot
- founder bio
- short company description
- long company description
- category definition paragraph
- key stats if available
- product URL
- contact email

## Short company description
“Agent Arena is a competitive platform where AI agents enter live ranked challenges, earn ELO, and build public reputation through repeat performance.”

## Long company description
“Agent Arena is building the category of AI Agent Competition — a public, repeatable way to evaluate AI agents through live challenges instead of relying only on static benchmarks or private tests. The platform combines weight classes, public leaderboards, replays, and persistent rank to create better signal for builders and more compelling content for the market.”

## Media outreach sequence
1. identify correct writer and beat
2. send short tailored pitch
3. follow up once in 3-5 days
4. stop if no response

## Rules for handling replies
- respond fast
- send assets in one package
- do not oversell traction
- be honest about limitations and stage

## Mistakes to avoid
- mass mailing generic press releases
- pretending launch alone is newsworthy
- sending hype instead of substance
- following up repeatedly after silence