AI Agents
Turn a Bot into a controlled worker
Define purpose, triggers, tools, boundaries, approval rules, and observable Runs for an AI Bot that performs repeatable work.
Where to open it
Path
Open Network > Bot Settings > Agent, or open Settings > AI & Automation > My Agents.
What it can do
- Trigger from Textxt messages, connected Gmail, supported business channels, schedules, or authenticated webhooks.
- Apply deterministic filters before an optional AI decision.
- Send a Textxt message, create a task, or prepare supported external replies under the configured approval policy.
Step by step
- Choose the AI Bot that supplies the Agent identity, provider, model, instructions, and Knowledge.
- Define one clear purpose, trigger, filters, AI decision instruction, and final action.
- Keep approval enabled for sensitive or external actions and verify the destination scope.
- Run dry tests, publish the version, then inspect Runs and Activity Log before expanding the workflow.
Ask an external AI
Paste this prompt into ChatGPT, Claude, or another assistant that can open public web pages:
Read https://textxt.com/help/ai-agents and guide me step by step. Use only capabilities documented on that page, ask for missing context, and do not invent buttons or settings.Example questions
- Create a Textxt Agent design that watches customer email and posts only messages that need a reply.
- Explain how approval, Runs, and Bot Knowledge work together in a Textxt Agent.
Good to know
- An AI Bot supplies the intelligence and identity; an Agent adds triggers, tools, boundaries, approvals, and execution history.
- Publishing creates a versioned workflow. Test a new draft before replacing a working production version.
- Textxt intentionally excludes arbitrary code and unbounded autonomous Agent graphs from the production safety boundary.
Keep private data private.
Never paste Textxt passwords, OAuth tokens, API keys, private chat exports, or confidential attachments into an external AI just to ask how the product works.