primitiv: Primitiv is a design contract layer for AI agents. It scans design sources like Figma, codebase, Storybook, and token files, reconciles conflicts between them, and exposes a single machine-readable contract via MCP. Agents connect to get authoritative answers before building UI, ensuring consistency and reducing drift.; agents-best-practices: A provider-neutral Agent Skill library for designing, auditing, and refactoring agentic harnesses compatible with Codex and Claude Code. It covers the full control plane of an agent runtime: typed tool design, permission checks, context management, memory, and observability. Targeted at developers building production-ready agent systems across any domain or AI provider.
AI agents retrieve the canonical design context before building UI components
Generate MVP agent harness blueprints for any business domain (CRM, ops, finance, healthcare)