Agent Claude Code
Implements Solana programs in Anchor 1.x: accounts, constraints, CPIs, errors, events, IDL and Rust LiteSVM tests. Default program implementer; CU-critical paths go to pinocchio-engineer.
Agent Claude Code
Integrates Solana DeFi protocols (Jupiter, Kamino, Marginfi, Raydium, Orca, Meteora, Sanctum, Pyth, Switchboard) via SDK, API or CPI. Token launches go to token-engineer.
Agent Claude Code
Owns CI/CD, toolchain pinning, RPC infrastructure, monitoring and Cloudflare Workers for Solana projects. Program deploys run through /deploy; service code goes to rust-backend-engineer.
Agent Claude Code
Designs Solana games before code: on-chain vs off-chain state, Unity architecture, economies, PSG1 targeting; writes concept.md and plan.md. Implementation goes to unity-engineer.
Agent Claude Code
Builds Solana mobile apps in React Native/Expo: Mobile Wallet Adapter, Seeker features, deep links, dApp Store releases. Web frontends go to solana-frontend-engineer.
Agent Claude Code
Implements CU- or size-critical Solana programs and hot paths in Pinocchio (zero-copy, manual validation, Mollusk CU tests) when Anchor is the measured bottleneck.
Agent Claude Code
Builds Rust async services around Solana (APIs, indexers, webhook consumers, transaction senders) on Tokio/Axum. Programs go to anchor-engineer; infra to devops-engineer.
Agent Claude Code
Designs and reviews Solana systems before code: accounts, PDAs, program boundaries, CPIs, upgrade and trust model, economic security. Games go to game-architect.
Agent Claude Code
Builds Solana web frontends in React/Next.js on @solana/kit: wallet connection, transaction UX, token views, WCAG 2.2 AA. React Native goes to mobile-engineer.
Agent Claude Code
Teaches Solana: explains concepts and existing code, writes tutorials and learning paths, mentors learners including EVM developers. Production code goes to engineer agents.
Agent Claude Code
Owns Solana test strategy and quality gates: Mollusk, LiteSVM, Surfpool and Trident suites, fuzzing, CU baselines, AI-slop cleanup. Implementers still unit-test their own code.
Agent Claude Code
Researches the Solana ecosystem with cited, dated sources: protocols, SDKs, APIs, tool versions, DeFi data, competitors. Design decisions go to solana-architect.
Agent Claude Code
Writes developer docs for Solana projects: READMEs, instruction and account references, integration and deployment guides. Tutorials go to solana-guide.
Agent Claude Code
Builds tokens and digital assets: SPL and Token-2022 mints and extensions, transfer-hook programs, Metaplex NFTs, launches, tokenomics. Swaps and lending go to defi-engineer.
Agent Claude Code
Implements Unity/C# for Solana games: Solana.Unity-SDK wallets, transactions, account reads, NFTs, UI, PSG1 input, Unity tests. Program changes go to anchor-engineer.
Command Claude Code
Audit infra security: secrets, deps, CI/CD, webhooks, AI/skill files.
Command Claude Code
Audit Solana program code for exploitable bugs and write a findings report.
Command Claude Code
Compare per-instruction CU with the stored baseline to catch regressions.
Command Claude Code
Build the web client (Next.js, Vite, React) and check env, types and bundle.
Command Claude Code
Build Solana programs (Anchor, Pinocchio, native), incl. verifiable builds.
Command Claude Code
Build the Unity project in batchmode for WebGL, desktop, Android or PSG1.
Command Claude Code
Turn a solana-ai-kit fork into a project: set up CLAUDE.md, remove kit files.
Command Claude Code
Un-ignore and commit the kit config (.claude/, CLAUDE.md, .mcp.json, .gitmodules).
Command Claude Code
Replay a user's failing transaction on forked state and map the error to source.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: