In SignalPilot, agents are your copilots inside the notebook. They understand your goals, write and edit code, and handle the routine work so you can focus on exploration and insight.

What Agents Do

• Interpret your requests – turn plain language into runnable code or actions.

• Generate & edit code – create new cells, refactor existing ones, or apply inline changes.

• Stay context-aware – remember your active datasets, database connections, and prior steps.

• Analyze & visualize – run computations, generate plots, and explain outputs in natural language.

• Debug intelligently – trace errors, explain stack traces, and propose fixes.

Why Agents Matter

Agents make SignalPilot more than just a notebook. They transform it into a collaborative environment, where the AI works alongside you as a partner—handling boilerplate, automating repetitive tasks, and guiding you through complex workflows.

Example Interaction

• You: “Query my database for last month’s sales and plot revenue by category.”

• SignalPilot Agent: connects to your DB, writes the SQL, runs the query, and creates the plot—all in one go.

FAQ

Q: What is an agent in SignalPilot?
An agent is an AI assistant inside the notebook that can generate code, edit cells, and interact with your data.

Q: How are SignalPilot agents different from other coding assistants?
Unlike static autocomplete, SignalPilot agents stay context-aware across cells, data, and database connections.

Q: Do I need to set up agents separately?
No. Agents are built into SignalPilot—just start chatting or asking for edits in your notebook. You just need to request an API key from us.

Q: Can agents remember my data and connections?
Yes, within the active session. SignalPilot agents keep track of context you provide, but follow a zero data retention policy outside your notebook.

Q: Can I use an agent for both coding and debugging?
Yes. SignalPilot agents generate, refactor, explain, and debug code—all in the same workflow.