Skip to content

Repository files navigation

cartesi-skills

Reusable skill packs for AI coding agents building with Cartesi Rollups.

What is in this repo

  • cartesi-workflow/ - starter guide and routing map; explains the phases of a full-stack Cartesi dApp and points to the right skill for each phase. Load this first when unsure where to start.
  • cartesi-scaffold/ - scaffold a new Cartesi Rollups v2 app from scratch; covers CLI version detection, cartesi create, and initial file layout.
  • cartesi-frontend/ - frontend-focused skill guidance for wallet, InputBox, JSON-RPC, inspect, and UI conventions.
  • cartesi-backend-core/ - language-agnostic backend core skill for advance_state, inspect_state, payload contracts, and finish-loop correctness.
  • cartesi-backend-py/ - Python-focused backend implementation skill.
  • cartesi-backend-js-ts/ - JavaScript/TypeScript-focused backend implementation skill.
  • cartesi-contracts/ - Solidity/Foundry skill for InputBox integrations, asset portals, voucher execution, and L1 wiring.
  • cartesi-local-dev/ - local and forked development workflow: build, run, advance/inspect, devnet tokens, and forked chain testing.
  • cartesi-deploy/ - deployment and operations skill for self-hosted environments; covers the full Docker Compose flow.
  • cartesi-jsonrpc/ - query a running node via the JSON-RPC 2.0 API (port 10011); full method reference with TypeScript types.
  • cartesi-debug/ - diagnose and fix errors across the full Cartesi stack, organised by layer and symptom.

Each skill is defined in a SKILL.md file and can include supporting docs (for example DESIGN.md).

How to use in AI agents

Most agent systems use a "skill" file as instruction context during generation.

  1. Pick the skill folder that matches the task.
  2. Load and include that folder's SKILL.md in the agent context.
  3. If present, also include supporting docs (for example DESIGN.md) as additional context.
  4. Ask the agent to implement changes in the target app repo while following the loaded skill guidance.

Suggested usage pattern

  • New to Cartesi / unsure where to start: include cartesi-workflow/SKILL.md first — it routes to the right skill for each phase.
  • Starting a new project: include cartesi-scaffold/SKILL.md.
  • Backend (language-agnostic core): include cartesi-backend-core/SKILL.md.
  • Backend (Python): include cartesi-backend-py/SKILL.md (plus backend core).
  • Backend (JS/TS): include cartesi-backend-js-ts/SKILL.md (plus backend core).
  • Frontend tasks: include cartesi-frontend/SKILL.md.
  • L1 contract tasks: include cartesi-contracts/SKILL.md.
  • Local development: include cartesi-local-dev/SKILL.md.
  • Deployment tasks: include cartesi-deploy/SKILL.md.
  • Querying node outputs programmatically: include cartesi-jsonrpc/SKILL.md.
  • Debugging errors or unexpected behaviour: include cartesi-debug/SKILL.md.
  • Custom UI requests: include cartesi-frontend/DESIGN.md to override default styling guidance.

Notes

  • Skills are guidance, not executable code.
  • Keep skills updated as Cartesi packages, APIs, and best practices evolve.

About

AI Agent Skills for Cartesi Application Development

Resources

Stars

1 star

Watchers

0 watching

Forks

Releases

Packages

Contributors