> ## Documentation Index
> Fetch the complete documentation index at: https://docs.rovax.io/llms.txt
> Use this file to discover all available pages before exploring further.

# Agent Settings

> Summary of all configuration sections available for an agent in Timely.ai and what each one is for.

Each agent has a set of settings you access by clicking on the agent in the list and navigating through the available tabs. Changes take effect immediately for new conversations — conversations already open are not affected.

## Summary

| Section                                                               | What it configures                                                                            |
| --------------------------------------------------------------------- | --------------------------------------------------------------------------------------------- |
| [Model](/en/guides/agents/settings/model)                             | LLM provider (OpenAI, Anthropic, Google, OpenRouter), specific model, temperature, and effort |
| [Internal name](/en/guides/agents/settings/internal-name)             | The agent's identifier within the platform — visible to the team, not to the customer         |
| [Role](/en/guides/agents/settings/role)                               | The agent's function and persona — the "position" it holds in the operation                   |
| [Prompt](/en/guides/agents/settings/prompt)                           | System instructions with AI editor, version history, and real-time quality analysis           |
| [Temperature](/en/guides/agents/settings/temperature)                 | Degree of creativity vs. consistency of responses (0.00 to 1.00)                              |
| [Effort](/en/guides/agents/settings/effort)                           | Reasoning depth controlled by maximum number of steps (10 to 100)                             |
| [Capabilities](/en/guides/agents/settings/capacities)                 | Optional features: planning, reasoning, automatic date and time                               |
| [Conditional prompts](/en/guides/agents/settings/conditional-prompts) | Instruction blocks activated by conversation or contact variables                             |
| [Internal chat](/en/guides/agents/settings/internal-chat)             | Direct testing interface with the agent without going through any external channel            |