Agents are the living center of Playnex — small, focused intelligences that help you think, write, research, plan, and create. They are not chatbots. They are collaborators. They have purpose, memory, and the ability to act.
Learn How Agents Work →An agent is a self‑contained intelligence with a purpose. It can think, plan, and act — using models, tools, and memory to help you accomplish work.
Unlike a chatbot, an agent is not a blank slate. It has:
Agents are the workers of the agent‑native era — quiet, capable, and always ready to help.
You can create an agent from the CLI:
playnex agent createYou’ll be asked to define:
This creates a new agent file in your workspace — a living definition that evolves over time.
Purpose is the most important part of an agent. It shapes how the model behaves, what it prioritizes, and how it interprets your requests.
Example purpose:
You are a research assistant.
You gather information, synthesize insights, and help me understand complex topics.
You write clearly, concisely, and with intellectual honesty.
Purpose is not static — you can refine it as your agent grows.
Tools extend what an agent can do. They allow agents to:
To add a tool:
playnex agent add-tool my-agentTools turn agents from thinkers into doers.
Agents can store information into capsules — drafts, notes, research, reflections, and more.
This gives them continuity. They can revisit past work, continue threads, and build long‑term understanding.
Memory is what makes agents feel alive — like collaborators who grow with you.
To talk to an agent:
playnex agent chat my-agentThis opens a conversational shell where the agent can think, plan, and act.
Agents can also run tasks, generate content, and save work into capsules automatically.
Agents are the foundation of Playnex — the workers, collaborators, and quiet intelligences that help you think and create.
Now that you understand agents, explore how memory works.
Or return to the documentation overview.
Back to Docs →