| Instructions Persistent context the agent sees every session — project conventions, coding standards, always-on guidance | | | | |
| Prompts Reusable prompt templates explicitly invoked as slash commands or by name | | | | |
| Skills Reusable workflows, knowledge, and prompt templates invoked on demand or auto-loaded when relevant | | | | |
| MCP & Tools Connections to external services, APIs, and databases via Model Context Protocol | | | | |
| Agents Specialist personas, isolated workers, or coordinated teams with distinct prompts and tools | | | | |
| Hooks Deterministic automation that runs on lifecycle events outside the LLM loop | | | | |
| Plugins & Distribution Packaging, sharing, and installing bundles of skills, hooks, agents, and MCP servers | | | | |
| Settings & Policy Tool configuration, managed settings, command rules, and organizational controls | | | | |