Capability Radar Launch Kit

Product name / brand

Capability Radar

Tagline

The weekly signal brief for people building useful AI agents.

One-line pitch

Capability Radar tracks MCP servers, Agent Skills, registries, payment protocols, and agent marketplaces — then tells AI agent builders what to test, skip, and watch.

Publishing channel

Primary channel: newsletter via Buttondown or Substack.
Public archive: GitHub Pages under Leonardo Research Labs.
Secondary distribution: X/Twitter thread, LinkedIn post, Hacker News / Reddit-style feedback post.

Brand decision

Do not use Enrico’s name. Use the standalone product brand Capability Radar, published under the existing Leonardo Research Labs technical publishing surface until/unless it deserves its own domain.

First CTA

Join the free weekly brief and send one agent capability you want tested.

Short CTA:

Get Capability Radar every week: 3 tools to test, 3 to skip, 3 to watch — for AI agent builders.

Launch post — X/Twitter

Agent builders have a new problem:

Too many MCP servers.
Too many agent skills.
Too many registries.
Too many “agent marketplaces.”
Too many payment protocols.
Not enough signal.

I’m launching Capability Radar: a weekly brief for builders that says what to test, skip, and watch.

Each issue covers:

  • MCP servers
  • Agent Skills / SKILL.md
  • agent tool registries
  • payment/access protocols
  • agent marketplaces
  • practical integration notes

No hype. No generic AI news. Just: “is this useful for builders or not?”

First issue: https://leodavinciad.github.io/capability-radar/issue-001.html

Reply with one agent capability you want evaluated.

Launch post — LinkedIn

AI agent builders are facing a tooling discovery problem.

Every week there are new MCP servers, Agent Skills, registries, marketplaces, and payment protocols. Some are useful. Some are duplicated. Some are not ready. Most are hard to evaluate quickly.

Capability Radar is a practical weekly brief for builders.

Each issue gives:

  • What to test this week
  • What to skip for now
  • What to watch
  • Integration notes
  • Risks and limitations
  • One recommended experiment

It is built for developers, AI automation builders, and technical founders who want signal, not another AI news feed.

Sample issue: https://leodavinciad.github.io/capability-radar/issue-001.html

If you are building with agents, reply with one tool, repo, MCP server, Agent Skill, or protocol you want reviewed in the first real issue.

Launch post — Hacker News / Reddit style

I’m testing a weekly brief for AI agent builders called Capability Radar.

The goal is simple: track new MCP servers, SKILL.md / Agent Skills, registries, payment/access protocols, and agent marketplaces, then summarize which ones are actually worth trying.

Format:

  • 3 things to test
  • 3 things to skip
  • 3 things to watch
  • one practical experiment
  • links + notes on maturity, docs, install friction, security, and usefulness

This is not meant to be a general AI newsletter. It is specifically for people building or operating agents.

Sample issue: https://leodavinciad.github.io/capability-radar/issue-001.html

Would this be useful? What sources/repos should it track?

14-day validation plan

Day 1

Create landing page/newsletter signup. Publish positioning post on X and LinkedIn.

Day 2

Manually DM/email 20 agent builders, MCP users, automation developers, and AI tool founders. Ask: “Would a weekly test/skip/watch brief save you time?”

Day 3

Publish sample mini-issue publicly with 5–9 evaluated items.

Day 4

Post the sample in relevant places:

  • X
  • LinkedIn
  • Hacker News / Reddit, if appropriate
  • Discords where MCP or agent builders gather, only if allowed

Day 5

Collect replies and requested tools. Track repeated categories.

Day 6

Ask 10 interested readers: “What would make this worth paying for: deeper tests, alerts, private database, integration recipes, or custom research?”

Day 7

Publish issue #001 / first official issue.

Week 2

Publish issue #002 with reader-submitted tools. Add a simple premium-interest CTA: “Want the full tested capability database and private monthly teardown? Reply premium.”

Validation metrics

Continue if:

  • 100+ landing page visits
  • 25+ email signups
  • 5+ direct replies
  • 10+ suggested tools/protocols
  • 3+ people say they would pay for deeper testing/database/access

Strong signal:

  • 1 sponsor inquiry
  • 1 paid custom evaluation request
  • 5+ premium-interest replies
  • builders share the issue without being asked