AI Agents
The signal tells you who. The AI agent makes it actionable.
- What are AI Agents and how do they workThe signal tells you who. The AI Agent makes it actionable. Agents read your social data, qualify it against criteria you define in plain language, and trigger the right action automatically — before a human touches it.
- Connecting an LLM provider (BYOK)AI Agents run on your own LLM account — you bring your own API key. This gives you full control over which models you use, how much you spend, and which provider you trust with your data. You must connect at least one provider before you can create or run AI Agents.
- Creating an AI Agent — Step 1: BuildThe Build step is where you define what your agent is, what data it processes, and which LLM powers it. Get these decisions right and the rest of the wizard follows naturally.
- Writing your AI Agent prompt — Step 2: PromptThe prompt is the intelligence of your agent. You write it in plain language, inject live social data using variable tags, and design the output to be simple enough to drive reliable conditional rules.
- Configuring actions — Step 3: ActionsActions are what your agent does with its output. Basic actions run every time. Conditional actions run only when the output matches a rule you define. Together they turn a classification or a piece of generated text into an automated workflow.
- Assigning AI Agents to SignalsAssigning an agent to a Signal means it runs automatically every time that Signal processes new data — no manual intervention needed. This is where AI Agents become a fully automated part of your workflow.
- Chaining AI AgentsChaining lets you pass the output of one agent directly into another, building multi-step workflows where each agent handles one focused task. The result is more reliable automation than trying to do everything in a single prompt.
- Running AI Agents on contacts manuallyYou can run any contact-targeting AI Agent against a selection of existing contacts directly from the Contacts manager — without waiting for a Signal to run.
- Prompt examples and templatesReady-to-use prompts for each data target. Copy, paste, and adapt to your ICP. Every example follows the same principle: simple, predictable output that drives reliable conditional rules.