> For the complete documentation index, see [llms.txt](https://doc.aissist.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://doc.aissist.io/capabilities.md).

# Capabilities

What Aissist's AI agents can do, and how each capability works.

*Last updated: October 25, 2026*

These pages explain what each part of Aissist is, why it matters and how it works. For step-by-step setup in the console, follow the [Tutorial](https://doc.aissist.io/tutorial.md).

### Build the agent

* [Instructions and Information](https://doc.aissist.io/capabilities/instructions-and-information.md) - the core idea: the AI's behavior comes from instructions (what to do) plus information (what it knows).
* [Sub Agents](https://doc.aissist.io/capabilities/sub-agents.md) - focused instructions, knowledge and actions for one scenario at a time.
* [Assets](https://doc.aissist.io/capabilities/assets.md) - the websites, pages, documents and help center content the AI answers from.
* [Actions](https://doc.aissist.io/capabilities/actions.md) - look up, create and update records in your systems during a conversation.
* [Connectors](https://doc.aissist.io/capabilities/connectors.md) - connect a system once per project and reuse it everywhere.
* [Languages](https://doc.aissist.io/capabilities/languages.md) - **65+ languages**, and how the AI picks the reply language.
* [Images and Media](https://doc.aissist.io/capabilities/multimedia.md) - **text, images, video and voice messages** customers send.

### Run it

* [Gateways](https://doc.aissist.io/capabilities/gateways.md) - put the agent to work in your help desk or chat platform.
* [Reliability](https://doc.aissist.io/capabilities/reliability.md) - how Aissist keeps errors rare and measures them.
* [Escalation and Handover](https://doc.aissist.io/capabilities/escalation-and-handover.md) - when the AI hands a conversation to your team, and how.
* [In-note Commands](https://doc.aissist.io/capabilities/in-note-command.md) - let your human agents ask the AI for help inside a ticket.

### Improve it

* [Evaluation](https://doc.aissist.io/capabilities/evaluation.md) - test the agent against realistic conversations before and after changes.
* [Continuous Improvement](https://doc.aissist.io/capabilities/continuous-improvement.md) - teach the AI from real conversations.
* [Ask Engine](https://doc.aissist.io/capabilities/ask.md) - the assistant in the console that answers questions about your workspace and sets things up for you.
* [Agent Insight](https://doc.aissist.io/capabilities/agent-insight.md) - what your conversations say about your customers and your agent.
* [Session Insight](https://doc.aissist.io/capabilities/session-insight.md) - extract the information you define from every closed conversation and write it to your help desk, so tickets stay up to date.
* [Insight Assets](https://doc.aissist.io/capabilities/insight-assets.md) - your QA guidelines and policies, used to check answers in Agent Insight, Evaluation, Session Insight and Analyze commands; never used to answer customers.
* [Weekly Reports and Alerts](https://doc.aissist.io/capabilities/weekly-reports.md) - a regular summary of results, plus alerts when something needs attention.

### Best practice

* Read [Instructions and Information](https://doc.aissist.io/capabilities/instructions-and-information.md) first. Every other page builds on it.
* Start with one workspace, a few assets and one or two sub-agents, then add actions and gateways.
* Use [Ask Engine](https://doc.aissist.io/capabilities/ask.md) whenever you are not sure where something belongs. It can explain and set up most of what these pages describe.

### Related

* [Quick Start](https://doc.aissist.io/tutorial/quick-start.md)
* [Success Metrics](https://doc.aissist.io/success-metrics.md)
