Is there a free plan?

Yes. The free tier includes ten hours of agent runtime with no credit card required, so you can build and deploy a working agent before paying anything.

Which AI models are supported?

Pinata Agents supports Claude (Anthropic), GPT (OpenAI), and any model exposed through OpenRouter — or skip API keys entirely with the built-in Pinata LLM and pay-as-you-go credits. You can switch models per agent at any time.

Can I bring my own API key?

Yes. Add your Anthropic, OpenAI, or OpenRouter key as an encrypted secret and your agent will use it for inference. Keys never leave your workspace and stay encrypted at rest.

Is self-hosting an option?

Yes — the agent runtimes are OpenClaw and Hermes, both open-source and self-hostable. Pinata Agents is the managed platform for running OpenClaw and Hermes — you skip Kubernetes setup, container builds, and TLS, and only pay for the time your agent is actually running.

How is this different from Replit Agent or Cursor?

Replit Agent and Cursor are coding assistants. Pinata Agents are persistent containerized agents that keep state across sessions, expose live web apps via routes, connect to messaging channels, and run scheduled tasks — closer to a managed Claude Code or Devin than an IDE plugin.

Can my agent run a real web app?

Yes. Any port your agent's container exposes can be mapped to a public URL via Routes. Up to 10 routes per agent, public or token-protected, served through Pinata's global edge with TLS and DDoS protection.

How do skills and the marketplace work?

Skills are pluggable capabilities — trading, DeFi monitoring, scraping, social posting — installed in one click from ClawHub. Each skill is a content-addressed package, so versions are pinned and reproducible.

Is everything versioned?

Yes. Every workspace change is captured as a content-addressed snapshot with a unique hash. You can browse history, diff between snapshots, and roll back to any prior state with one click.

Can my agent work on issues autonomously?

Yes. Every agent has an issue board — file an issue, assign it to your agent, and it implements the change on its own branch and opens a pull request. Super Builder agents also report per-run cost and a full file-by-file diff for review before you merge.

Yes. Install the agents CLI with a single command and manage everything from your terminal — create and deploy agents, send prompts, stream logs, run tasks, and manage secrets. Everything the dashboard does is scriptable.

onchainos — Deploy this AI agent in one click on Pinata

onchainos — OpenClaw Agent Template

Built for AI. Ready for Web3.

A general-purpose OpenClaw agent template powered by OKX OnchainOS. Deploy on any OpenClaw-compatible host and get an on-chain research and trading agent that can research tokens, track smart money, scan new launches, analyse wallets, manage DeFi positions, and execute swaps across 500+ aggregated DEXs on 60+ networks.

Quick start

Clone or fork this template
Run bash setup.sh — installs the onchainos CLI and fetches skills + workflows from the source repo
Start an OpenClaw session — the agent is ready

To enable trading, run onchainos wallet login inside a session.

What you can do

Ask the agent	What it does
"Research this token: `<address>`"	Price, security scan, holders, smart money signals, dev reputation
"What is smart money buying?"	Aggregated SM/KOL/whale signals with per-token due diligence
"Scan new tokens on pump.fun"	MIGRATED token list with safety, dev reputation, and bundle analysis
"Analyse this wallet: `<address>`"	7d/30d PnL, trading behaviour, recent on-chain activity
"Daily brief"	Market prices, hot tokens, SM activity, new launches
"Check my portfolio: `<address>`"	Balances, total value, PnL overview
"Buy 0.1 SOL of BONK"	Pre-trade risk detection → quote → confirm → MEV-protected execution

Infrastructure

500+ DEX sources aggregated for best swap price
130+ networks via OKX Wallet ecosystem
Sub-100ms average response times · 99.9% uptime
TEE-secured agentic wallet — private keys never exposed
MEV protection — Jito (Solana) and Flashbots (EVM)
Gas-free payments on X Layer via OKX Agent Payments Protocol

Structure

openclaw_template/
├── README.md                     # This file
├── manifest.json                 # Template store manifest
├── setup.sh                      # Build script (calls install.sh)
└── workspace/
    ├── SOUL.md                   # Agent personality, values, tone, boundaries
    ├── AGENTS.md                 # Workflow routing, skill table, harness rules, session management
    ├── BOOTSTRAP.md              # First-run onboarding (self-deletes after setup)
    ├── IDENTITY.md               # Agent name, type, vibe, emoji
    ├── USER.md                   # Learned user preferences (updated by agent)
    ├── TOOLS.md                  # Capabilities, CLI reference, wallet modes, swap infrastructure
    ├── MEMORY.md                 # Long-term learned patterns (updated by agent)
    ├── HEARTBEAT.md              # Periodic task config
    ├── memory/                   # Daily memory files (memory/YYYY-MM-DD.md)
    └── projects/
        └── scan-bot-example.py   # Sample Python script demonstrating onchainos CLI scripting

Skills and workflows are fetched from the onchainos-skills source repo at deploy time — always the latest version.

Optional: OKX API credentials

The agent uses built-in sandbox keys by default (rate-limited). For production-grade rate limits, set these environment variables:

Variable	Description
`OKX_API_KEY`	Apply at OKX Developer Portal
`OKX_SECRET_KEY`	Your API secret
`OKX_PASSPHRASE`	Your API passphrase

Links

License

MIT

onchainos

Required Setup

Optional