solana-founder-brand

A personalized X/Twitter personal brand playbook skill for Solana founders — grounded in real Solana-founder archetypes, not generic content-marketing advice.

Install

npx skills add danielAsaboro/solana-founder-brand

Works with Claude Code, Cursor, Codex, Gemini CLI, and 15+ other agents. Installs to .agents/skills/solana-founder-brand/ and symlinks to each agent's skills directory automatically.

What it does

Give the skill your name, what you're building, your background, and what you care about. It:

1. Matches you to one of 8 researched Solana-founder archetypes (or a blend).
2. Derives your unique POV — the one perspective you can own on X.
3. Builds a community identity layer: a name for your early users, roles, and how to make them feel like co-owners.
4. Produces 3–4 content pillars, each with a core CTA (the specific action each post drives readers to take).
5. Writes a voice style guide and weekly posting rhythm.
6. Names your engagement playbook: real Solana X handles, regional Telegram/Discord seeding targets, trending-conversation insertion tactics.
7. Generates 10 example posts in your voice — including at least one community post structure (Early Adopter Spotlight, Community Naming Moment, or Proof-of-Use).

Outputs: a filled playbook.md, an importable content-calendar.csv (Typefully / Buffer / Hypefury), and a machine-readable engagement-playbook.json.

Usage

After installing, just ask in plain English inside any supported agent:

Build me a founder brand playbook for my Solana project.
I'm building [what]. Background: [your background].
I care about [what matters to you]. My X handle: @yourhandle

Or run non-interactively from a project directory:

cd your-project
claude -p "Build me a founder brand playbook. I'm building [X]. Background: [Y]. Care about: [Z]." \
  --model sonnet \
  --dangerously-skip-permissions

The skill picks up your context from the conversation — name, project, background, communication style, and optionally your X handle for live voice grounding.

MCP server (optional — upgrades to native tool calls)

The mcp/ directory contains an MCP server that exposes fetch_x_posts, web_search, and analyze_voice as native Claude Code tools instead of Bash subprocess calls. Without it the skill falls back to the scripts/ equivalents automatically.

Setup:

# 1. Clone the repo (the installed skill doesn't include the MCP server)
git clone https://github.com/danielAsaboro/solana-founder-brand.git
cd solana-founder-brand

# 2. Install MCP server dependencies
cd mcp && npm install && cd ..

# 3. Copy env template and fill in your keys
cp .env.example .env
# Edit .env — TWITTERAPI_IO_KEY and TAVILY_API_KEY are the two that matter most

# 4. Register in your Claude Code project settings (.claude/settings.local.json or .mcp.json)
# The repo ships a .mcp.json with relative paths — Claude Code picks it up automatically.

The .mcp.json in the repo root works out of the box once you have the .env filled:

{
  "mcpServers": {
    "solana-founder-brand": {
      "command": "node",
      "args": ["--env-file=.env", "./mcp/server.mjs"]
    }
  }
}

Environment variables

All optional — the skill works without any keys using the archetype library and paste fallback. Each key upgrades one capability.

Without keys the skill still produces a complete playbook — it just uses the archetype library for voice matching instead of the founder's real posts.

How to report a bug

Open an issue at github.com/danielAsaboro/solana-founder-brand/issues.

License

MIT