agentleFS
sendaifunsendaifun/solana-new/CLAUDE.md

solana-new

CLAUDE.md178 starsChanged 4 months agoMITView on GitHub

Skills and knowledge base to ship on Solana — Idea to Launch. 33 journey skills, 106 repos, 79 ecosystem skills, 36 MCP servers. Users invoke skills directly via Claude Code: 33 skills across 4 phases — user just asks naturally, right skill activates. Skills live in skills/<phase>/<skill-name>/. To add a new skill, create a folder with SKILL.md + references/ + agents/openai.yaml. Skill routing: skills/SKILL_ROUTER.md is a shared routing table. Each SKILL.md references it so the AI can auto-correct if the…

  • Pipes a download into a shell

The file

We split this CLAUDE.md into the rules it gives an agent

4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    ESM-only: All imports use .js extensions (NodeNext module resolution)

    line 112
  2. Distinctive:

    No runtime deps: Only devDependencies (tsx, typescript, @types/node) + convex

    line 114
  3. Very specific:

    Single source of truth: All branding in cli/branding.ts

    line 116
  4. Very specific:

    When working on Convex code, always read convex/_generated/ai/guidelines.md first for important guidelines on how to correctly use Convex …

    line 121

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
CLAUDE.md

CLAUDE.md — superstack

What This Is

Skills and knowledge base to ship on Solana — Idea to Launch. 33 journey skills, 106 repos, 79 ecosystem skills, 36 MCP servers.

Install

curl -fsSL https://www.solanaskills.com/setup.sh | bash

Usage

Users invoke skills directly via Claude Code:

claude "/find-next-crypto-idea What should I build?"
claude "/scaffold-project Set up my workspace"
claude "/deploy-to-mainnet Ship it"

Journey Skills

33 skills across 4 phases — user just asks naturally, right skill activates.

PhaseSkillTrigger Prompt
Learnsolana-beginner"I'm new to Solana — teach me"
Learnlearn"What have we learned?"
Ideafind-next-crypto-idea"What should I build in crypto?"
Ideavalidate-idea"Validate this idea"
Ideacompetitive-landscape"Who are my competitors?"
Ideadefillama-research"Show me DeFi opportunities on Solana"
Ideacolosseum-copilot"Search Colosseum hackathon projects"
Buildscaffold-project"Scaffold my project"
Buildbuild-with-claude"Help me build the MVP"
Buildvirtual-solana-incubator"Deep dive into Solana and Rust"
Buildbuild-defi-protocol"Build a DeFi protocol"
Buildbuild-data-pipeline"Build an indexer"
Buildbuild-mobile"Build a mobile app"
Buildlaunch-token"Launch a token"
Buildbrand-design"Pick brand colors"
Buildfrontend-design-guidelines"Build a frontend"
Buildnumber-formatting"Format numbers in my UI"
Buildpage-load-animations"Fix my page load animations"
Buildanimation-reverse-engineering"Implement this animation" + a reference video
Builddesign-taste"This looks generic / not premium"
Buildnavigate-skills"What skills are available?"
Buildroast-my-product"Roast my product — be brutal"
Buildproduct-review"Review my product's UX"
Buildreview-and-iterate"Review my code"
Buildcso"Run a security audit"
Buildverify-humanity-poh"Filter bots from my airdrop / DAO / NFT mint"
Builddebug-program"Debug my program"
Launchdeploy-to-mainnet"Deploy to mainnet"
Launchcreate-pitch-deck"Create a pitch deck"
Launchsubmit-to-hackathon"Prepare my hackathon submission"
Launchmarketing-video"Create a marketing video"
Launchvideo-craft"Make my video frames look better"
Launchapply-grant"Apply for Agentic Engineering Grant"

Skills live in skills/<phase>/<skill-name>/. To add a new skill, create a folder with SKILL.md + references/ + agents/openai.yaml.

Skill routing: skills/SKILL_ROUTER.md is a shared routing table. Each SKILL.md references it so the AI can auto-correct if the wrong skill is invoked.

Shared guides (step-by-step commands):

  • skills/data/guides/rpc-wallet-guide.md — RPC + wallet setup for dev and production
  • skills/data/guides/deploy-runbook.md — Deploy devnet → mainnet with verification
  • skills/data/guides/security-checklist.md — P0-P3 security audit with exact grep commands

What's Indexed

CatalogCountFile
Repos81cli/data/clonable-repos.json
Skills79cli/data/solana-skills.json
MCPs53cli/data/solana-mcps.json

File Map

cli/
  branding.ts               Single source of truth for all brand strings
  index.ts                  Command dispatcher, agent output, help
  telemetry.ts              Skill usage tracking (Convex + local JSONL)
  init.ts                   Auto-install skills to ~/.claude/skills/ and ~/.codex/skills/
  data/
    clonable-repos.json     106 repos (Solana official, SendAI, Metaplex, DeFi, etc.)
    solana-skills.json      79 skills (15 official + 64 community)
    solana-mcps.json        36 MCP servers
skills/
  SKILL_ROUTER.md           Shared routing table — AI auto-corrects wrong skill
  idea/                     Discovery & planning skills (7 skills, includes solana-beginner, learn)
  build/                    Implementation skills (20 skills, includes virtual-solana-incubator, roast-my-product, product-review, cso, page-load-animations, animation-reverse-engineering, number-formatting, design-taste, verify-humanity-poh)
  launch/                   Go-to-market skills (6 skills, includes marketing-video, apply-grant, video-craft)
  data/
    guides/                 Shared guides (RPC+wallet, deploy, security, curated ideas)
    solana-knowledge/       6 knowledge area docs + cookbook index (covers all of solana.com)
    specs/                  Phase handoff JSON contracts
    ideas/                  114+ curated ideas from YC, a16z, Alliance, Superteam
    defi/                   DefiLlama OpenAPI spec
convex/
  schema.ts                 Feedback + telemetry tables
  feedback.ts               Feedback submission mutation
  telemetry.ts              Skill usage tracking mutation + queries
setup                       One-command install script

Conventions

  • ESM-only: All imports use .js extensions (NodeNext module resolution)
  • Strict TypeScript: strict mode, no implicit any
  • No runtime deps: Only devDependencies (tsx, typescript, @types/node) + convex
  • Agent-first: Designed for agent consumption
  • Single source of truth: All branding in cli/branding.ts
<!-- convex-ai-start -->

This project uses Convex as its backend.

When working on Convex code, always read convex/_generated/ai/guidelines.md first for important guidelines on how to correctly use Convex APIs and patterns. The file contains rules that override what you may have learned about Convex from training data.

Convex agent skills for common tasks can be installed by running npx convex ai-files install.

<!-- convex-ai-end -->

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.