AI Engineering

AuraKit is a fullstack development engine for Claude Code that adds 34 command modes, OWASP Top 10 security rules, runtime hooks, and token reduction. Use /aura to build, fix, review, or deploy code across multiple languages and frameworks.

Implement Feature reads a feature specification, designs architecture, writes code and tests, and prepares a structured handoff to QA. Use it when an approved feature spec is ready for implementation by the Forja (Dev) agent.

SDD Pipeline collects rule and schema reference documents used by pipeline agents (specifier, planner, scheduler, builder, verifier, committer). It is intended as an internal reference for agents, not for direct user invocation.

Tutti Agent Workspace App builds or evolves complex agent-enabled Tutti app repositories with web/server/shared monorepos, @tutti-os/agent-acp-kit local runtimes, dynamic agent catalogs, and run-scoped MCP tool gateways. Use it for full Tutti app architecture, packaging, and production validation; for simple package creation use tutti-workspace-app-factory instead.

Claude API provides guidance for building against Anthropic APIs and SDKs, including Messages, streaming, prompt caching, tool use, batches, and files. Use it when integrating Claude into applications or agent workflows.

ClawHub CLI searches, installs, updates, and publishes agent skills from clawhub.com. Use it to fetch new skills on the fly, sync installed skills to latest or a specific version, or publish new/updated skill folders.

Codebase Exploration uses SocratiCode semantic search, dependency graphs, and context artifacts to help an agent understand code structure, locate functions and types, and analyze dependencies when exploring code, including database schemas and API specs.

Coding Agent delegates coding tasks to Codex, Claude Code, or Pi agents via background process. Use it for building features, reviewing PRs, refactoring large codebases, or iterative coding that requires file exploration.

LibTV Image & Video generates AI images and videos through Seedance 2.0 and other models via the LibTV gateway. Use it for text-to-image, image-to-video, video continuation, style transfer, and text-to-video tasks.

Mondo Poster Design generates AI image prompts and actual designs in Mondo's bold screen-print aesthetic, useful for movie posters, book covers, album art, and event posters.

Nano Banana generates or edits images using Nano Banana image models (nano-banana, nano-banana-pro, nano-banana-2). Use it when you need to create an image from a prompt, edit an existing image, or compose multiple images into a collage.

Nano Banana One Shop generates and edits images using the Nano Banana model family (Gemini 3.1 Flash, 3 Pro, and 2.5 Flash). Use it to create images from prompts, edit existing images, or compose multiple images into one output.