Solana Claude Configuration

Production-ready Claude Code configuration for full-stack Solana development. Combines best practices from multiple sources into an agent-optimized, token-efficient config you can copy over and adapt to your specific project.

Current 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.

The idea here is to provide a generic CLAUDE.md that relies on subagents to plan and execute actions, dynamically loading markdown files for context, saving tokens in the end of the day.

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.

What This Is

A complete .claude/ configuration that turns Claude into a Solana development expert with:

• Specialized agents for different tasks (architecture, Anchor, Pinocchio, frontend, backend, QA, docs, Unity games)
• Workflow commands for building, testing, deploying, and committing (Rust, TypeScript, Unity/C#)
• Progressive skill loading that only loads context when needed (saves tokens)
• Auto-loading rules that enforce best practices based on file patterns

Quick Start

# Copy to your Solana project
cp -r .claude /path/to/your-project/
cp CLAUDE-solana.md /path/to/your-project/CLAUDE.md

# Or clone and use as template
git clone https://github.com/solanabr/solana-claude-config

Key Features

Agent-Ready Architecture

Each agent loads its own specialized context on invocation:

"Use solana-architect to design the vault program"
"Use anchor-engineer to implement the deposit instruction"  
"Use solana-qa-engineer to write comprehensive tests"

Claude will spawn each specialized agent by itself based on context.

Token-Efficient Design

• Skills load progressively (not all at once)
• Agents reference skills instead of duplicating content
• Rules auto-load only for matching file patterns
• Decision frameworks live in agents, not global context

Modern Stack (2026)

Repository Structure

.
├── CLAUDE.md                    # Main hub - Claude reads this first
├── README.md                    # This file
├── LICENSE                      # MIT
└── .claude/
    ├── agents/                      # 11 specialized agents
    │   ├── solana-architect.md          # System design, PDAs, architecture
    │   ├── anchor-engineer.md           # Anchor framework development
    │   ├── pinocchio-engineer.md        # CU optimization, zero-copy
    │   ├── solana-frontend-engineer.md  # React/Next.js, wallet UX
    │   ├── rust-backend-engineer.md     # Axum APIs, indexers
    │   ├── solana-qa-engineer.md        # Testing, CU profiling, code quality
    │   ├── tech-docs-writer.md          # Documentation
    │   ├── game-architect.md            # Solana game design, Unity architecture
    │   ├── unity-engineer.md            # Unity/C# implementation, Solana.Unity-SDK
    │   ├── solana-guide.md              # Learning, tutorials, concept explanations
    │   └── solana-researcher.md         # Ecosystem research, protocol investigation
    │
    ├── commands/                # 14 workflow commands
    │   ├── quick-commit.md          # Branch creation, format, lint, commit
    │   ├── build-program.md         # Build Solana programs
    │   ├── build-app.md             # Build web clients
    │   ├── build-unity.md           # Build Unity projects (WebGL, Desktop, PSG1)
    │   ├── test-rust.md             # Rust tests (Mollusk/LiteSVM/Trident)
    │   ├── test-typescript.md       # TypeScript tests
    │   ├── test-dotnet.md           # .NET/C# tests (Unity, NUnit)
    │   ├── test-and-fix.md          # Automated test→fix loop
    │   ├── deploy.md                # Deploy (devnet first, then mainnet)
    │   ├── audit-solana.md          # Security audit workflow
    │   ├── setup-ci-cd.md           # GitHub Actions setup
    │   ├── write-docs.md            # Generate documentation
    │   ├── explain-code.md          # Explain complex code with diagrams
    │   └── plan-feature.md          # Plan feature implementation
    │
    ├── skills/                   # Progressive-loading knowledge
    │   ├── SKILL.md                  # Entry point, routing guide
    │   ├── programs-anchor.md        # Anchor patterns
    │   ├── programs-pinocchio.md     # Pinocchio patterns
    │   ├── frontend-framework-kit.md # @solana/kit patterns
    │   ├── kit-web3-interop.md       # Legacy web3.js interop
    │   ├── backend-async.md          # Axum/Tokio patterns
    │   ├── testing.md                # Test framework selection
    │   ├── security.md               # Security checklist
    │   ├── idl-codegen.md            # IDL generation and client codegen
    │   ├── payments.md               # Solana Pay, Commerce Kit, Kora payments
    │   ├── deployment.md             # Deployment workflows
    │   ├── ecosystem.md              # DeFi/NFT integrations
    │   ├── resources.md              # Official links
    │   ├── unity.md                  # Solana.Unity-SDK, C# patterns
    │   └── playsolana.md             # PlaySolana ecosystem, PSG1, PlayDex
    │
    ├── rules/                   # Auto-loading constraints
    │   ├── anchor.md                # Anchor code rules
    │   ├── pinocchio.md             # Pinocchio code rules
    │   ├── rust.md                  # General Rust rules
    │   ├── kit-react.md             # Frontend rules
    │   └── dotnet.md                # C#/.NET/Unity rules
    │
    └── settings.json            # Permissions, hooks

Agents

Commands

GitHub Action for Team Collaboration

This config works with the Claude GitHub Action for PR-based iteration:

1. Add ANTHROPIC_API_KEY to your repository secrets
2. Team members can @claude in PR comments
3. Claude responds with code suggestions using this configuration

Branch Workflow

All new work starts on a feature branch:

# Format: <type>/<scope>-<description>-<DD-MM-YYYY>
git checkout -b feat/program-vault-15-01-2026
git checkout -b fix/frontend-auth-15-01-2026

Use /quick-commit to automate branch creation and commits.

Code Quality

Before merging, check for AI-generated patterns that don't match the codebase:

git diff main...HEAD

Remove: excessive comments, abnormal try/catch blocks, verbose errors, redundant validation.

Keep: legitimate security checks, non-obvious explanations, matching error patterns.

Credits

This project builds on excellent work from the community:

• GuiBibeau/solana-dev-skill - Gui from the Solana Foundation open-sourced a comprehensive skill set for Solana development. This project restructures those skills for better agent coordination and progressive loading.

• 0xquinto/bcherny-claude - Compiled Boris Cherny's (creator of Claude Code at Anthropic) best practices including verification loops, parallel Claude sessions, and CLAUDE.md patterns.

• builderz-labs/solana-claude-md - Provided starter inspiration for Solana-specific Claude configuration.

License

MIT - See LICENSE

TODO

This is a work in progress, use at your own risk.

• Integrate any MCP tools used in project
• Wait for bug to be solved and review rules/skills token usage
• Test pinnochio usage