Aomi

> Aomi for AI agents — drive on-chain action and build new app integrations through natural-language prompts. Works with Claude Code, Cursor, Gemini CLI, OpenAI Codex, VS Code Copilot, and any Agent Skills–compatible AI tool.

What's in this bundle

The two skills are different audiences and different risk profiles. Bundling means the agent loads only the skill whose description triggers match — there is no token bloat for installing both.

Install

Via Claude Code (after Anthropic listing lands)

/plugin marketplace add anthropics/claude-code
/plugin install aomi

Via Self-hosted marketplace (canonical alternative)

/plugin marketplace add aomi-labs/skills
/plugin install aomi

Via gh extension

gh extension install ai-ecoverse/gh-upskill
gh upskill aomi-labs/skills --skill aomi

Via direct clone

git clone https://github.com/aomi-labs/skills
cp -r skills/plugins/aomi ~/.claude/skills/

Prerequisites

# Required for aomi-transact (the on-chain CLI driver)
npm install -g @aomi-labs/client@latest      # v0.1.42 or newer

aomi-build has no runtime prerequisites — it generates Rust source code from specs.

Usage

Once installed, ask your agent in natural language. The agent picks the right skill based on the prompt:

Triggers aomi-transact:

• "What's the price of ETH?"
• "Swap 1 USDC for WETH on Uniswap V3, send to my wallet."
• "Stake 0.01 ETH with Lido."
• "Bridge 50 USDC from Ethereum to Base via CCTP."

Triggers aomi-build:

• "Use aomi-build to turn this OpenAPI spec into an Aomi app."
• "Build an Aomi plugin from these REST endpoints."
• "Convert this SDK README into an Aomi tool surface."

Bundle structure

aomi/
├── .claude-plugin/
│   └── plugin.json             # bundle manifest
├── README.md                   # this file
├── LICENSE                     # MIT
├── SECURITY.md                 # bundle-level OWASP AST01–AST10 walkthrough
└── skills/
    ├── transact/
    │   ├── SKILL.md            # main skill
    │   ├── agents/openai.yaml  # Codex/OpenAI host metadata
    │   ├── references/
    │   │   ├── account-abstraction.md
    │   │   ├── apps.md
    │   │   ├── drain-vectors.md
    │   │   ├── examples.md
    │   │   ├── thread.md
    │   │   └── troubleshooting.md
    │   └── templates/
…

Security posture

aomi-transact ships with a complete OWASP AST03 (Over-Privileged Skills) permission manifest in its frontmatter — risk_tier: L2, permissions.{files,network,shell,tools} declarations. aomi-build is risk_tier: L1 (scaffolds code, no transaction side effects).

The bundle has been scanned by four independent tools: Cisco AI Defense skill-scanner, pors/skill-audit, NMitchem/SkillScan, and Snyk agent-scan. Captured reports live at .scanner-reports/ in the repo root. Per-skill OWASP AST01–AST10 walkthroughs are in SECURITY.md.

License

MIT. See LICENSE.

About Aomi

Aomi Labs builds native harness around blockchains functioning like Claude Code on-chain. We specialize in executions against arbitrary protocol with non-custodial workflow, account abstraction, and full security with simulations. Aomi also host agentic applications deployed and owned by developers, companies, and agents. Aomi provides E2E integration with UI, Skills and SDKs.

Links:

• 🌐 Website: aomi.dev
• 🤖 Agents: aomi.dev/agents
• 𝕏 Twitter: x.com/aomi_labs
• 💻 GitHub: github.com/aomi-labs
• 📦 Packages:
• @aomi-labs/react
• aomi-sdk