By kocaemre
AI agent skills for building confidential dApps with the Zama Protocol — context7-aware, anti-hallucination
Run an FHE-aware code review on a Solidity + TypeScript codebase using @fhevm/solidity. Detects four high-impact issue classes — missing ACL grants (FHE.allowThis / FHE.allow), cleartext leaks via require/event/emit on decrypted values, HCU-explosion functions (>12 warning, >20 error FHE ops in one tx), and imports of officially deprecated packages (fhevm root, fhevmjs). Produces AUDIT-REPORT.md with severity classification.
One-command end-to-end pipeline. Runs design → init → contract → test → audit → deploy → frontend in sequence, pausing only at safety gates (audit findings, the disable-model-invocation deploy step, frontend smoke test). Best for first-time users who want a working confidential dApp without thinking about which skill comes next.
Author confidential Solidity contracts using @fhevm/solidity (euint, ebool, eaddress, ACL, FHE.allowThis). Use when the user wants to write or modify FHE-aware smart contracts, integrate OpenZeppelin Confidential Contracts (ERC-7984, governance), or pick a decryption path (public/user/oracle).
Diagnose fhEVM / Zama Protocol errors from a pasted revert message, vitest failure, hardhat trace, or runtime exception. Matches the input against a curated catalog of 10 known patterns (ACL grants, deprecated fhevmjs/fhevm imports, HCU exhaustion, relayer SSR/indexedDB, Etherscan v1, relayer 502, wagmi ABI drift, ZamaConfig path, /bundle vs /web entry) and returns a likely root cause, concrete fix command(s), and an authoritative reference link.
Deploy compiled fhEVM contracts to Sepolia testnet, verify on Etherscan, and (if applicable) auto-register with the Confidential Token Registry. Use ONLY when the user explicitly asks to deploy.
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Zama Protocol skill pack for Claude Code — from an empty directory to a working confidential dApp in 30 minutes.
Built for: Zama Developer Program — Mainnet Season 2 / Bounty Track.
Differentiator: Every generated line is verified live against official Zama documentation via context7 MCP — /zama-ai/fhevm (1,772 snippets), /zama-ai/fhevm-hardhat-template, /websites/openzeppelin_confidential-contracts. Zero hallucinated APIs.
| Tool | Why | Install |
|---|---|---|
| Node.js >= 20 | Runtime for skill scripts | nvm → nvm install 20 |
| pnpm 10+ | Package manager (skill-generated projects use it) | npm install -g pnpm@10 |
context7 MCP | REQUIRED — live Zama / OZ Confidential docs (anti-hallucination) | claude mcp add context7 -- npx -y @upstash/context7-mcp |
magic MCP (21st.dev) | RECOMMENDED — better UI scaffolds for /zama-frontend and /zama-design | claude mcp add magic -- npx -y @21st-dev/magic |
After install, verify with /zama-doctor — it checks every requirement and prints fix commands for whatever is missing.
/plugin marketplace add github.com/kocaemre/zama-skills
/plugin install zama-skills@zama-skills
/zama-doctor # confirm Node/pnpm/context7/magic are ready
The npm package ships an interactive multi-tool installer. Run it in your project root:
npx zama-skills install
You'll get a checklist:
zama-skills installer
Select every AI tool you want the skill rules installed for.
❯ ◉ Claude Code (detected) Native plugin (slash commands, auto-routing)
◉ Cursor Rules under .cursor/rules/zama-skills/
◯ OpenCode Rules + AGENTS.md pointer
◯ Codex CLI (OpenAI) AGENTS.md convention
◯ Aider CONVENTIONS.md pointer
◯ Continue (VS Code) .continue/rules/
◯ Generic (any AI tool) zama-skills-knowledge/ + README
Tools you already use are pre-detected. Press SPACE to toggle, ENTER to confirm. The installer drops the right asset format into the right directory and (where supported) appends a pointer block to your master rules file (AGENTS.md, CONVENTIONS.md).
Non-interactive flags for CI:
npx zama-skills install --tool cursor,opencode --force
npx zama-skills install --all --force # install for every supported tool
--scope project (default) writes under cwd — the rules ship with your repo, your team gets them on git pull. This is the right choice for almost every tool.
--scope personal writes under $HOME — only Claude Code actually reads global rules from ~/.claude/skills/. The other tools key off the current working directory.
| Tool | Default location (--scope project) | --scope personal works? |
|---|---|---|
| Claude Code | <cwd>/.claude/skills/zama-skills/ | ✅ yes — ~/.claude/skills/ is real |
| Cursor | <cwd>/.cursor/rules/zama-skills/ | ❌ project-local only |
| OpenCode | <cwd>/.opencode/rules/zama-skills/ + <cwd>/AGENTS.md | ❌ project-local only |
| Codex CLI | <cwd>/.codex/rules/zama-skills/ + <cwd>/AGENTS.md | ❌ project-local only |
| Aider | <cwd>/.aider/zama-skills/ + <cwd>/CONVENTIONS.md | ❌ project-local only |
| Continue | <cwd>/.continue/rules/zama-skills/ | ❌ project-local only |
| Generic | <cwd>/zama-skills-knowledge/ | ✅ either works (you wire it up) |
The CLI prints a warning if you pick --scope personal for a tool that won't honor it. Recommended: keep the default --scope project, commit the rule files to your repo, and let your team's AI agents pick them up automatically.
▶ Watch the 2-minute walkthrough · Live dApp · Verified contract
npx claudepluginhub kocaemre/zama-skills --plugin zama-skillsUI/UX design intelligence. 67 styles, 161 palettes, 57 font pairings, 25 charts, 15 stacks (React, Next.js, Vue, Svelte, Astro, SwiftUI, React Native, Flutter, Tailwind, shadcn/ui, Nuxt, Jetpack Compose). Actions: plan, build, create, design, implement, review, fix, improve, optimize, enhance, refactor, check UI/UX code. Projects: website, landing page, dashboard, admin panel, e-commerce, SaaS, portfolio, blog, mobile app. Elements: button, modal, navbar, sidebar, card, table, form, chart. Styles: glassmorphism, claymorphism, minimalism, brutalism, neumorphism, bento grid, dark mode, responsive, skeuomorphism, flat design. Topics: color palette, accessibility, animation, layout, typography, font pairing, spacing, hover, shadow, gradient.
Comprehensive skill pack with 66 specialized skills for full-stack developers: 12 language experts (Python, TypeScript, Go, Rust, C++, Swift, Kotlin, C#, PHP, Java, SQL, JavaScript), 10 backend frameworks, 6 frontend/mobile, plus infrastructure, DevOps, security, and testing. Features progressive disclosure architecture for 50% faster loading.
This skill should be used when users need to generate ideas, explore creative solutions, or systematically brainstorm approaches to problems. Use when users request help with ideation, content planning, product features, marketing campaigns, strategic planning, creative writing, or any task requiring structured idea generation. The skill provides 30+ research-validated prompt patterns across 14 categories with exact templates, success metrics, and domain-specific applications.
Develop, test, build, and deploy Godot 4.x games with Claude Code. Includes GdUnit4 testing, web/desktop exports, CI/CD pipelines, and deployment to Vercel/GitHub Pages/itch.io.
Upstash Context7 MCP server for up-to-date documentation lookup. Pull version-specific documentation and code examples directly from source repositories into your LLM context.
A growing collection of Claude-compatible academic workflow bundles. Covers scientific figures, manuscript writing and polishing, reviewer assessment, citation retrieval, data availability, paper reading, literature search, response letters, paper-to-PPTX conversion, and evidence-grounded Chinese invention patent drafting. Rules are organized as reusable skill folders with explicit workflows and quality checks.