{"id":51223198,"url":"https://github.com/solanabr/solana-ai-kit","last_synced_at":"2026-07-01T04:00:36.170Z","repository":{"id":332688754,"uuid":"1133771838","full_name":"solanabr/solana-ai-kit","owner":"solanabr","description":"Claude Code / Codex / AI configs for the expert Solana builder. CLAUDE.md, agents, commands, hooks, rules, skills and settings across Web, Anchor, Pinnochio, Unity, Mobile, security and more.","archived":false,"fork":false,"pushed_at":"2026-06-26T22:35:09.000Z","size":4460,"stargazers_count":87,"open_issues_count":27,"forks_count":49,"subscribers_count":1,"default_branch":"main","last_synced_at":"2026-06-27T00:32:35.264Z","etag":null,"topics":["claude-agent","claude-code","claude-commands","claude-md","claude-skills","claude-subagents","solana-agents","solana-ai","solana-claude","solana-engineer","solana-program"],"latest_commit_sha":null,"homepage":"https://superteam.codes","language":"Shell","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/solanabr.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":null,"dco":null,"cla":null}},"created_at":"2026-01-13T19:52:26.000Z","updated_at":"2026-06-26T22:35:09.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/solanabr/solana-ai-kit","commit_stats":null,"previous_names":["solanabr/solana-claude-config","solanabr/solana-claude","solanabr/solana-ai-kit"],"tags_count":8,"template":true,"template_full_name":null,"purl":"pkg:github/solanabr/solana-ai-kit","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/solanabr%2Fsolana-ai-kit","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/solanabr%2Fsolana-ai-kit/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/solanabr%2Fsolana-ai-kit/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/solanabr%2Fsolana-ai-kit/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/solanabr","download_url":"https://codeload.github.com/solanabr/solana-ai-kit/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/solanabr%2Fsolana-ai-kit/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":34992071,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-07-01T02:00:05.325Z","response_time":130,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["claude-agent","claude-code","claude-commands","claude-md","claude-skills","claude-subagents","solana-agents","solana-ai","solana-claude","solana-engineer","solana-program"],"created_at":"2026-06-28T09:00:31.118Z","updated_at":"2026-07-01T04:00:36.157Z","avatar_url":"https://github.com/solanabr.png","language":"Shell","funding_links":[],"categories":["Source Catalog"],"sub_categories":[],"readme":"\u003cp align=\"center\"\u003e\u003cimg src=\".github/assets/solana-ai-kit-banner.png\" alt=\"Solana AI Kit — Skill, MCP \u0026 config aggregator for Claude Code, Codex + any agentic setup\" width=\"100%\" /\u003e\u003c/p\u003e\n\n# Solana AI Kit\n\n[![CI](https://github.com/solanabr/solana-ai-kit/actions/workflows/ci.yml/badge.svg)](https://github.com/solanabr/solana-ai-kit/actions/workflows/ci.yml)\n![Version](https://img.shields.io/badge/version-2.0.0-blue)\n![License](https://img.shields.io/badge/license-MIT-blue)\n![Solana](https://img.shields.io/badge/Solana-black?logo=solana)\n![Claude Code](https://img.shields.io/badge/Claude_Code-powered-orange)\n\nProduction-ready Claude Code configuration for full-stack Solana development. Combines best practices from multiple sources into an agent-optimized, token-efficient config you can install and adapt to your specific project.\n\nThe idea here is to provide a generic CLAUDE.md that relies on subagents to plan and execute actions, dynamically loading markdown files, saving tokens and context in the end of the day. This config fully leverages the official Claude Code config recommendations:\n- Rules are only loaded whenever specific file types are involved;\n- SKILL.md is a mega hub to dynamically-disclosed skill files that are directly fetched from the best skill repos distributed across the ecosystem (Solana Foundation, Colosseum, Solana Mobile, SendAI, etc);\n- Plus, its CLAUDE-solana.md is than half the size of the usual CLAUDE.md, leaving space for its self improvements programmed into the agents, noting and learning from anti-patterns, errors, recurrency and more. For less important notes, CLAUDE.local.md is constantly maintained by agents as well and, on monorepos, per-folder CLAUDE.md is also maintained.\n\nCurrent multi-agent workflow favors monorepos, so we use a single CLAUDE.md/config for the whole project while leveraging agents and context-specific skills to solve each step of builder flow.\n\nIf you installed manually, remember to rename ./CLAUDE-solana.md back to ./CLAUDE.md, as the current top-level CLAUDE.md file is focused on maintaining the repo itself.\n\n## What This Is\n\nA complete `.claude/` configuration that turns Claude into a Solana development expert with:\n\n- **15 specialized agents** for different tasks (architecture, Anchor, Pinocchio, DeFi, tokens, frontend, mobile, backend, DevOps, QA, docs, games, Unity, learning, research)\n- **30 workflow commands** for building, testing, deploying, profiling, migrating, and committing\n- **7 MCP server integrations** for on-chain data (Helius), Solana docs (solana-dev), library docs (Context7), browser automation (Playwright), context optimization (context-mode), persistent memory (memsearch), and local-validator / mainnet-fork control (Surfpool)\n- **Agent teams** for multi-step workflows (architect → engineer → QA)\n- **Progressive skill loading** that only loads context when needed (saves tokens)\n- **Auto-loading rules** that enforce best practices based on file patterns\n\n## Quick Start\n\n```bash\n# Option 0: Fork Template (Github UI)\nclaude -m \"/cleanup\"  # then start claude code running /cleanup so top-level directory of your project isn't polluted\n\n\n# Option 1: One-liner installer (Claude Code)\ncurl -fsSL https://aikit.superteam.codes | bash\n# Fallback (or before the aikit.superteam.codes DNS is live): curl -fsSL https://raw.githubusercontent.com/solanabr/solana-ai-kit/main/install.sh | bash\n\n# Option 2: One-liner installer (Codex, Opencode, everything else)\ncurl -fsSL https://aikit.superteam.codes | bash -s -- --agents\n# Fallback (or before the aikit.superteam.codes DNS is live): curl -fsSL https://raw.githubusercontent.com/solanabr/solana-ai-kit/main/install.sh | bash -s -- --agents\n\n# Option 3: Manual setup\ngit clone --recurse-submodules https://github.com/solanabr/solana-ai-kit.git\ncp -r solana-ai-kit/.claude /path/to/your-project/\ncp solana-ai-kit/CLAUDE-solana.md /path/to/your-project/CLAUDE.md\ncd /path/to/your-project \u0026\u0026 git submodule update --init --recursive\n\n# Option 4: Claude Code plugin (run inside Claude Code)\n/plugin marketplace add solanabr/solana-ai-kit\n/plugin install solana-ai-kit@stbr\n# Commands then namespace as /solana-ai-kit:\u003cname\u003e. The plugin is the CORE kit\n# (agents/commands/local skills/MCP/hooks); the curl one-liner (Option 1) is the\n# FULL install, adding rules + permissions/sandbox + the 18 ext/ submodules.\n# Details: see \"Install as a Claude Code plugin\" below.\n\n# Start Claude Code\nclaude\n```\n\n### `--agents` Flag\n\nPass `--agents` to install everything into `.agents/` instead of `.claude/`. Same content, different directory name. Useful when `.claude/` is already taken or for non-Claude AI tools (Cursor, Windsurf, Copilot, etc.).\n\n```bash\nbash install.sh --agents /path/to/your-project\n```\n\nTo update an `--agents` install:\n\n```bash\n# Run from your project root:\nbash .agents/bin/update.sh\n```\n\n### Config is gitignored by default\n\nTo keep your project clean, the installer adds `.claude/`, `CLAUDE.md`, `.mcp.json`, and `.gitmodules` to `.gitignore` — the kit reads as ignorable infrastructure, not your app code (the `ext/` skill submodules are ignored too, re-fetched via `git submodule update`).\n\nWant the config tracked in git (team setup, reproducible config)? Run `/commit-claude-config` — it un-ignores those files and commits them (or edit `.gitignore` by hand). If your project already commits `.claude/` or its own `.gitmodules`, the new ignore lines are a no-op — git keeps tracking files it already tracks.\n\n### MCP Setup (Optional)\n\nAfter installation, configure MCP servers for enhanced capabilities:\n\n```bash\n# In your project with Claude Code running:\n/setup-mcp\n```\n\nThis guides you through API key configuration for Helius, Context7, and other MCP servers.\n\n## Install as a Claude Code plugin\n\nsolana-ai-kit is also its own Claude Code marketplace serving one **core plugin**. From inside Claude Code:\n\n```text\n/plugin marketplace add solanabr/solana-ai-kit\n/plugin install solana-ai-kit@stbr\n```\n\nThe plugin ships the **core kit**: the 15 agents, 30 commands, the local go-to-market + registry skills (idea-sprint, pitch-deck, hackathon), the 7 MCP servers, and the dev hooks (banner, formatter, pre-deploy/commit gates). Commands and skills are namespaced — `/deploy` becomes `/solana-ai-kit:deploy`.\n\nWhat the plugin **cannot** carry (Claude Code plugins are plain git clones — they can't init submodules or ship a permissions/sandbox policy), so these stay exclusive to the **full install** (`install.sh`):\n\n- the lazy-loaded `.claude/rules/*` code-style law (Rust, Anchor, Pinocchio, TypeScript, .NET)\n- the curated permissions allowlist + sandbox policy\n- the 18 `ext/` skill submodules (protocol, security, infra, ecosystem depth)\n\nFor protocol-skill depth in plugin form, add the upstream marketplaces instead (routing, not copying) — e.g. `/plugin marketplace add sendaifun/skills`, `ghostsecurity/skills`, `trailofbits/skills`, `cloudflare/skills`. The plugin's skill hub and `skill-registry.json` list the current targets.\n\nThe two paths are complementary: individuals wanting Solana agents/commands across **all** their projects → plugin at user scope; project teams working in one repo → `install.sh` (full kit). If you enable both in the same project, `/doctor` warns about duplicate commands/hooks/MCP and advises picking one. Plugin auto-update is off by default for third-party marketplaces — refresh with `/plugin marketplace update` + `claude plugin update`.\n\n## External Skill Submodules\n\n| Submodule | Source | Purpose |\n|-----------|--------|---------|\n| `ext/solana-dev` | [solana-foundation/solana-dev-skill](https://github.com/solana-foundation/solana-dev-skill) | Core Solana development (programs, frontend, testing, security) |\n| `ext/sendai` | [sendaifun/skills](https://github.com/sendaifun/skills) | DeFi protocol integrations (Jupiter, Raydium, Kamino, perps, cross-chain, oracles, etc.) |\n| `ext/solana-game` | [solanabr/solana-game-skill](https://github.com/solanabr/solana-game-skill) | Game development (Unity, PlaySolana, PSG1) |\n| `ext/cloudflare` | [cloudflare/skills](https://github.com/cloudflare/skills) | Infrastructure (Workers, Agents SDK, MCP servers) |\n| `ext/trailofbits` | [trailofbits/skills](https://github.com/trailofbits/skills) | Security auditing and vulnerability scanning |\n| `ext/qedgen` | [QEDGen/solana-skills](https://github.com/QEDGen/solana-skills) | Formal verification with Lean 4 theorem proving |\n| `ext/solana-mobile` | [solana-mobile/solana-mobile-dev-skill](https://github.com/solana-mobile/solana-mobile-dev-skill) | Mobile Wallet Adapter, Genesis Token, SKR address resolution |\n| `ext/colosseum` | [ColosseumOrg/colosseum-copilot](https://github.com/ColosseumOrg/colosseum-copilot) | Startup research, idea validation, hackathon projects |\n| `ext/safe-solana-builder` | [frankcastleauditor/safe-solana-builder](https://github.com/frankcastleauditor/safe-solana-builder) | Security-first code generation (70+ audit-derived rules) |\n| `ext/vercel` | [vercel-labs/agent-skills](https://github.com/vercel-labs/agent-skills) | Vercel deployment, Next.js, AI SDK, v0, edge functions |\n| `ext/solana-new` | [sendaifun/solana-new](https://github.com/sendaifun/solana-new) | 32 idea→launch journey skills + idea datasets/knowledge base; routed via local wrappers |\n| `ext/ghostsecurity` | [ghostsecurity/skills](https://github.com/ghostsecurity/skills) | 7 AppSec skills: SAST criteria, SCA, secrets, validation |\n| `ext/defending-code` | [anthropics/defending-code-reference-harness](https://github.com/anthropics/defending-code-reference-harness) | Anthropic vuln-discovery reference harness + 6 skills |\n| `ext/jupiter` | [jup-ag/agent-skills](https://github.com/jup-ag/agent-skills) | Official Jupiter skills: Ultra swap, Lend, swap migration, VRFD |\n| `ext/metaplex` | [metaplex-foundation/skill](https://github.com/metaplex-foundation/skill) | Official Metaplex: Core, Token Metadata, Bubblegum, Candy Machine, Genesis |\n| `ext/helius` | [helius-labs/core-ai](https://github.com/helius-labs/core-ai) | Official Helius infra skill + unique SVM internals skill |\n| `ext/quicknode-anchor` | [quiknode-labs/solana-anchor-claude-skill](https://github.com/quiknode-labs/solana-anchor-claude-skill) | Anchor/financial-math/Quasar reference files (quarantined — refs only) |\n| `ext/eth-to-sol` | [solana-foundation/eth-to-sol-skill](https://github.com/solana-foundation/eth-to-sol-skill) | EVM/Solidity → Anchor two-pass porting |\n\n### Agent Teams\n\nEach agent loads its own specialized context on invocation:\n\n```\n\"Use solana-architect to design the vault program\"\n\"Use anchor-engineer to implement the deposit instruction\"\n\"Use defi-engineer to integrate Jupiter swaps\"\n\"Use solana-qa-engineer to write comprehensive tests\"\n```\n\nClaude will spawn each specialized agent by itself based on context and, with `CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS` enabled, orchestrate multi-agent workflows:\n\n```\n\"Create an agent team: solana-architect for design, anchor-engineer for implementation, solana-qa-engineer for testing\"\n```\n\nRecommended team patterns:\n\n| Pattern | Flow | Use Case |\n|---------|------|----------|\n| **program-ship** | architect → anchor/pinocchio → qa | Build program from spec to tested |\n| **full-stack** | architect → anchor → frontend → qa | End-to-end feature |\n| **audit-and-fix** | qa → trailofbits context → anchor | Audit and remediate |\n| **game-ship** | game-architect → unity → qa | Game feature |\n| **defi-compose** | researcher → defi-engineer → qa | DeFi integration |\n| **token-launch** | token-engineer → frontend → qa | Token creation + launch UI |\n\n### MCP Server Integrations\n\nPre-configured MCP servers in `.mcp.json` (API keys go in `.env`):\n\n| Server | Capabilities |\n|--------|-------------|\n| **Helius** | 60+ tools: RPC, DAS API, webhooks, priority fees, token metadata, NFT data |\n| **solana-dev** | Solana Foundation official MCP — Solana docs, guides, and API references |\n| **Context7** | Up-to-date library documentation lookup |\n| **Playwright** | Browser automation for dApp testing and visual verification |\n| **context-mode** | Context window optimization — compresses large RPC responses, build logs, and code analysis |\n| **memsearch** | Persistent AI memory across sessions — auto-captures summaries, semantic search, git-friendly storage |\n| **Surfpool** | Agent-driven local validator / mainnet-fork control via the Surfpool CLI — `surfpool mcp` (requires the surfpool CLI) |\n\n### Token-Efficient Design\n\n- CLAUDE.md is delivered as a user message (not system prompt) — shorter = better adherence\n- Skills load progressively (not all at once)\n- Agents reference skills instead of duplicating content\n- Path-scoped rules lazy-load on file read (zero startup cost)\n- `CLAUDE.local.md` for private scratch notes (gitignored, never shared)\n- Subdirectory CLAUDE.md files lazy-load in monorepos\n- Decision frameworks live in agents, not global context\n\n### Modern Stack (2026)\n\n| Layer | Stack |\n|-------|-------|\n| Programs | Anchor 1.0+, Pinocchio, Rust 1.82+ |\n| Token Extensions | Token-2022 (transfer hooks, confidential transfers, metadata) |\n| Testing | Mollusk, LiteSVM, Surfpool, Trident |\n| Frontend | @solana/kit, Next.js 15, React 19 |\n| Mobile | React Native, Expo, Mobile Wallet Adapter |\n| Backend | Axum 0.8+, Tokio 1.40+, sqlx |\n| Unity Games | Solana.Unity-SDK, .NET 9, C# 13 |\n| PlaySolana | PSG1 console, PlayDex, SvalGuard |\n| DeFi | Jupiter, Drift, Kamino, Raydium, Orca, Meteora |\n| Infrastructure | Cloudflare Workers, GitHub Actions, Docker |\n\n## Extended Skills \u0026 Add-on Registry\n\nBeyond the bundled submodules above, the kit ships a curated catalog of **opt-in** skills, plugins, and MCPs that are **not installed by default** — the agent installs them only on your request, at your own expense. The set is deduped against [solana-new's](https://github.com/sendaifun/solana-new) vendored ecosystem catalogs so it surfaces net-new, high-signal picks rather than restating what's already reachable.\n\nFeatured add-ons by domain:\n\n- **Claude-official:** [anthropics/skills](https://github.com/anthropics/skills) · [anthropics/claude-code](https://github.com/anthropics/claude-code) plugins (non-OSI license; overlaps `/diff-review` + `cso`)\n- **Dev-workflow:** [wshobson/agents](https://github.com/wshobson/agents) · [gsd-build/get-shit-done](https://github.com/gsd-build/get-shit-done)\n- **Frontend/Design:** [zarazhangrui/frontend-slides](https://github.com/zarazhangrui/frontend-slides) · [uxKero/anydesign](https://github.com/uxKero/anydesign) · [dylantarre/animation-principles](https://github.com/dylantarre/animation-principles)\n- **UX/Writing:** [content-designer/ux-writing-skill](https://github.com/content-designer/ux-writing-skill) · [cuellarfr/design-skills](https://github.com/cuellarfr/design-skills)\n- **Testing:** [SawyerHood/dev-browser](https://github.com/SawyerHood/dev-browser) · [conorluddy/ios-simulator-skill](https://github.com/conorluddy/ios-simulator-skill)\n- **Data:** [K-Dense-AI/scientific-agent-skills](https://github.com/K-Dense-AI/scientific-agent-skills) · [Nansen](https://github.com/nansen-ai/nansen-cli) (paid analytics MCP)\n\n**Where we scout** new tools (aggregators, not installable): [ComposioHQ/awesome-claude-skills](https://github.com/ComposioHQ/awesome-claude-skills) · [travisvn/awesome-claude-skills](https://github.com/travisvn/awesome-claude-skills) · [davepoon/buildwithclaude](https://github.com/davepoon/buildwithclaude) · [hesreallyhim/awesome-claude-code](https://github.com/hesreallyhim/awesome-claude-code) · [VoltAgent/awesome-claude-code-subagents](https://github.com/VoltAgent/awesome-claude-code-subagents).\n\nFor broader Solana coverage, see solana-new's vendored catalogs at `ext/solana-new/cli/data/` (MCPs, skills, clonable repos).\n\nSee [`skill-registry.json`](.claude/skills/skill-registry.json) for the complete extended catalog — every entry with its install command, license, and safety caveats.\n\n## Repository Structure\n\n```\n.\n├── CLAUDE.md                    # Main hub - Claude reads this first\n├── README.md                    # This file\n├── .mcp.json                    # MCP server configurations (project root)\n├── install.sh                   # One-liner installer\n├── update.sh                    # Deprecation wrapper → .claude/bin/update.sh\n├── validate.sh                  # Config integrity checker\n├── LICENSE                      # MIT\n├── tests/                       # Config integrity test suite\n├── .github/workflows/\n│   ├── ci.yml                       # PR validation\n│   ├── claude-code.yml              # Claude Code action template\n│   └── claude.yml                   # @claude mention responder (issues/PRs)\n└── .claude/\n    ├── VERSION                  # Semver version (e.g. 2.0.0)\n    ├── agents/                  # 15 specialized agents\n    ├── bin/\n    │   ├── update.sh                # In-place update from upstream\n    │   └── resync.sh                # Submodule resync script\n    ├── commands/                # 30 workflow commands\n    ├── skills/                  # Progressive-loading knowledge\n    │   ├── SKILL.md                 # Unified hub routing to all skills\n    │   ├── ext/                     # External skill submodules\n    │   │   ├── solana-dev/              # Solana Foundation dev skill (core)\n    │   │   ├── sendai/                  # SendAI protocol skills (DeFi)\n    │   │   ├── solana-game/             # Solana game skill (Unity, PSG1)\n    │   │   ├── cloudflare/              # Cloudflare Workers, Agents SDK\n    │   │   ├── trailofbits/             # Trail of Bits security skills\n    │   │   ├── qedgen/                # QEDGen formal verification (Lean 4)\n    │   │   ├── solana-mobile/           # Mobile Wallet Adapter, Genesis Token\n    │   │   ├── colosseum/              # Colosseum Copilot (startup research)\n    │   │   ├── safe-solana-builder/   # Security-first code generation\n    │   │   ├── vercel/                # Vercel deployment, Next.js, AI SDK\n    │   │   ├── solana-new/            # SendAI idea→launch journey skills + datasets\n    │   │   ├── ghostsecurity/         # Ghost Security AppSec skills\n    │   │   ├── defending-code/        # Anthropic vuln-discovery reference harness\n    │   │   ├── jupiter/               # Official Jupiter skills (swap, lend, VRFD)\n    │   │   ├── metaplex/              # Official Metaplex (NFT, candy machine)\n    │   │   ├── helius/                # Official Helius infra + SVM internals\n    │   │   ├── quicknode-anchor/      # Anchor/Quasar reference files (quarantined)\n    │   │   └── eth-to-sol/            # EVM/Solidity → Anchor porting\n    │   ├── skill-registry.json     # Opt-in add-on catalog (repos/skills/MCPs)\n    │   ├── idea-sprint/             # Wrapper: find + validate crypto ideas (GTM)\n    │   ├── pitch-deck/              # Wrapper: pitch decks for crypto projects (GTM)\n    │   ├── hackathon/               # Wrapper: hackathon submissions + grants (GTM)\n    │   ├── token-2022.md            # Token Extensions guide (local)\n    │   ├── backend-async.md         # Axum/Tokio patterns (local)\n    │   └── deployment.md            # Deployment workflows (local)\n    ├── rules/                   # Auto-loading constraints\n    └── settings.json            # Permissions, hooks, agent teams\n```\n\n## Agents\n\n| Agent | Purpose | Model |\n|-------|---------|-------|\n| **solana-architect** | System design, PDA schemes, token economics, multi-program architecture | Opus |\n| **anchor-engineer** | Anchor development, IDL generation, account constraints | Opus |\n| **pinocchio-engineer** | CU optimization (80-95% savings), zero-copy, minimal binary | Opus |\n| **defi-engineer** | DeFi integrations: Jupiter, Drift, Kamino, Raydium, Orca, Meteora | Opus |\n| **token-engineer** | Token-2022 extensions, token economics, launch mechanics | Opus |\n| **solana-frontend-engineer** | React/Next.js, wallet UX, transaction flows, accessibility | Opus |\n| **mobile-engineer** | React Native/Expo, mobile wallet adapter, deep linking | Sonnet |\n| **rust-backend-engineer** | Axum APIs, indexers, WebSocket services | Opus |\n| **devops-engineer** | CI/CD, monitoring, RPC infrastructure, Cloudflare Workers | Sonnet |\n| **solana-qa-engineer** | Testing (Mollusk/LiteSVM/Trident), CU profiling, code quality | Opus |\n| **tech-docs-writer** | READMEs, API docs, integration guides | Sonnet |\n| **game-architect** | Solana game design, Unity architecture, on-chain game state, PlaySolana | Opus |\n| **unity-engineer** | Unity/C# implementation, Solana.Unity-SDK, wallet integration, NFT display | Sonnet |\n| **solana-guide** | Learning, tutorials, concept explanations, progressive learning paths | Sonnet |\n| **solana-researcher** | Ecosystem research, protocol investigation, SDK analysis | Sonnet |\n\n## Commands\n\n### Building\n| Command | Purpose |\n|---------|---------|\n| `/build-program` | Build Anchor or native Solana program |\n| `/build-app` | Build Next.js/Vite web client |\n| `/build-unity` | Build Unity project (WebGL, Desktop, PSG1) |\n| `/scaffold` | Generate project scaffolding (program + frontend + tests + CI) |\n\n### Testing \u0026 Quality\n| Command | Purpose |\n|---------|---------|\n| `/test-rust` | Run Mollusk, LiteSVM, Surfpool, Trident tests |\n| `/test-ts` | Run TypeScript tests (Anchor, Vitest, Playwright) |\n| `/test-dotnet` | Run .NET/C# tests (Unity Test Framework, NUnit) |\n| `/test-and-fix` | Run tests and auto-fix common issues |\n| `/audit-solana` | Comprehensive security audit |\n| `/audit-infra` | Infra-first security audit: secrets, supply chain, CI/CD, LLM/skill security, OWASP, STRIDE |\n| `/product-review` | Product quality review — UX walkthrough, 8-dimension scorecard (`--harsh` for the roast variant) |\n| `/diff-review` | AI-powered diff review for Solana-specific issues |\n| `/profile-cu` | CU profiling per instruction with optimization suggestions |\n| `/benchmark` | CU benchmarks with before/after comparison |\n| `/debug-user-tx` | Replay a user's failing tx against forked state, map error to source |\n\n### Deployment \u0026 Migration\n| Command | Purpose |\n|---------|---------|\n| `/deploy` | Deploy to devnet (always first) or mainnet |\n| `/migrate-web3` | Migrate from @solana/web3.js to @solana/kit |\n| `/generate-idl-client` | Generate typed clients from IDL (Codama/Shank) |\n\n### Workflow \u0026 Setup\n| Command | Purpose |\n|---------|---------|\n| `/quick-commit` | Create branch, format, lint, conventional commit |\n| `/commit-claude-config` | Version the kit config in git (un-ignores `.claude/`, `CLAUDE.md`, `.mcp.json`, `.gitmodules`) |\n| `/setup-ci-cd` | Configure GitHub Actions pipeline |\n| `/setup-mcp` | Configure MCP server API keys and connections |\n| `/resync` | Resync external skill submodules to latest |\n| `/write-docs` | Generate documentation for programs, APIs, components |\n| `/explain-code` | Explain complex code with visual diagrams |\n| `/plan-feature` | Plan feature implementation with specifications |\n| `/update` | Update config to latest version from upstream |\n| `/cleanup` | Initialize forked template — setup CLAUDE.md, remove scaffolding |\n| `/doctor` | Health check for dev environment + config — read-only, one fix-it command per failure |\n| `/dream` | Memory consolidation — dedupe, prune, re-rank MEMORY.md and Project Learnings |\n\n## DX Scripts\n\n| Script | Purpose |\n|--------|---------|\n| `install.sh` | One-liner installer: copies config to your project (`--agents` for non-Claude tools) |\n| `update.sh` | Deprecation wrapper → `.claude/bin/update.sh` |\n| `validate.sh` | Validates all config integrity (agents, commands, skills, settings, versioning) |\n| `tests/run_all.sh` | Runs full test suite for config validation |\n\n## GitHub Action for Team Collaboration\n\nThis config includes a pre-built GitHub Action (`.github/workflows/claude-code.yml`) for PR-based iteration:\n\n1. Add `ANTHROPIC_API_KEY` to your repository secrets\n2. Copy `.github/workflows/claude-code.yml` to your project\n3. Team members can `@claude` in PR comments\n4. Claude responds with code suggestions using this configuration\n\n## Branch Workflow\n\nAll new work starts on a feature branch:\n\n```bash\n# Format: \u003ctype\u003e/\u003cscope\u003e-\u003cdescription\u003e-\u003cDD-MM-YYYY\u003e\ngit checkout -b feat/program-vault-15-01-2026\ngit checkout -b fix/frontend-auth-15-01-2026\n```\n\nUse `/quick-commit` to automate branch creation and commits.\n\n## Code Quality\n\nBefore merging, run `/diff-review` or check diff against main:\n\n```bash\ngit diff main...HEAD\n```\n\nRemove: excessive comments, abnormal try/catch blocks, verbose errors, redundant validation.\n\nKeep: legitimate security checks, non-obvious explanations, matching error patterns.\n\n## Using as a GitHub Template\n\n1. Click \"Use this template\" on GitHub (or fork the repo)\n2. Clone your new repo\n3. Run Claude Code and use `/cleanup` — this copies `CLAUDE-solana.md` → `CLAUDE.md` and removes config repo scaffolding (tests, install scripts, docs)\n4. Start building!\n\nFor monorepos, add a `CLAUDE.md` to each package/module with architecture decisions scoped to that directory. Claude Code automatically loads these when working in that subdirectory.\n\n## Updating\n\n```bash\n# Preferred: use /update in Claude Code, or run directly:\nbash .claude/bin/update.sh\n\n# Preview changes without applying:\nbash .claude/bin/update.sh --dry-run\n\n# Resync skill submodules only:\nbash .claude/bin/resync.sh\n\n# Or manually update submodules\ngit submodule update --remote --merge\n```\n\n## License\n\nMIT - See [LICENSE](LICENSE)\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsolanabr%2Fsolana-ai-kit","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsolanabr%2Fsolana-ai-kit","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsolanabr%2Fsolana-ai-kit/lists"}