EARLY ACCESS --- ## Before you begin Keep these guidelines in mind throughout the configuration: - Use the same language across [knowledge sources](https://www.infobip.com/docs/agentos-ai-agents/knowledge-agents/connect-knowledge-sources) and prompts. - Be consistent with brand name, agent purpose, and knowledge source content. Inconsistency causes the agent to give contradictory responses. - For prompt writing best practices, see [Write prompts for AI agents](https://www.infobip.com/docs/agentos-ai-agents/advanced-topics/write-prompts). NOTE If you upload documentation for product A but write the prompt *You should be a helpful agent for product B*, the agent gives conflicting answers. ___ ## Create the agent NOTE You can create a maximum of **3 knowledge agents** per account. If you reach the limit, the **Create knowledge agent** button is unavailable. To create a new agent, [delete an existing one](https://www.infobip.com/docs/agentos-ai-agents/knowledge-agents/manage-knowledge-agents#delete-a-knowledge-agent) first. 1. On the Infobip web interface, go to **AI Agents** > **My agents** > **Knowledge agents** tab. 2. Select **Create knowledge agent**. 3. Enter a unique name for the agent. This name is not shown to end users. 4. In the **Agent** tab, [configure the agent settings](#configure-the-agent-settings). 5. In the **Knowledge base** tab, [connect knowledge sources](https://www.infobip.com/docs/agentos-ai-agents/knowledge-agents/connect-knowledge-sources). 6. Select **Save agent**. ___ ## Configure the agent settings In the **Agent** tab, configure the following fields. | Field | Description | |-------|-------------| | **Description** | Describe what the agent does, for example, *This agent answers customer questions about our product using the company knowledge base*. This description is for internal reference only and is not shown to end users. | | **Instructions** | Define the agent behavior, persona, boundaries, and response style. The instructions tell the agent how to respond to end users. Include: persona, response behavior, forbidden topics, and example conversations. | **Example** ``` - You are a helpful knowledge assistant for [COMPANY]. - Answer questions based on the provided documentation. - Be concise, accurate, and friendly. - If the answer is not in the documentation, say so honestly — do not make up facts. - Support follow-up questions using conversation context. - Do not answer questions about: [competitors, politics, or other off-topic subjects]. - Example: "I'm sorry, I can only help with questions about [COMPANY] products and services." - Example conversation: User: What are your business hours? Assistant: Based on our documentation, we're open Monday to Friday, 9 AM to 5 PM. User: Are you open on weekends? Assistant: No, we're currently closed on weekends. ``` ___ ## Response settings Configure how the agent generates responses. | Setting | Description | |---|---| | **Session window** | The number of recent messages from the current conversation included as context when generating a response. A larger window provides more continuity; a smaller window reduces processing overhead. Minimum: 0. Maximum: 10. | | **Retrieved chunks** | The number of knowledge base chunks retrieved and passed to the model as context. A higher number provides broader context; a lower number keeps responses more focused. Range: 1–4. | | **Output tokens** | The maximum number of tokens the model can use in a single response. This setting limits response length only, use prompting to control answer style and conciseness. Minimum: 1. Maximum: 512. | | **Temperature** | Controls how predictable or varied the model's responses are. Lower values produce consistent, focused answers. Higher values produce more varied responses but may reduce factual precision. Range: 0–2. | IMPORTANT Avoid high temperature values. The agent may produce incorrect or unclear responses. ___ ## Index settings Configure how the knowledge base content is indexed. | Setting | Description | |---|---| | **Paragraph size** | The size of each content chunk in the knowledge base index. Larger values capture more context per chunk; smaller values create more granular retrieval. Minimum: 1. Maximum: 900. | | **Overlapping tokens** | The number of tokens shared between adjacent chunks in the knowledge base index. Increasing overlap reduces the chance that relevant content is split across chunk boundaries. Minimum: 0. Maximum: 50. | ___ ## Safety Configure content filters to control what the agent can and cannot discuss. Guardrails detect harmful content in user messages. Select one or more guardrails to enable: - **Hate** - **Self harm** - **Violence** - **Sexual** - **Jailbreak shield** For more details on guardrail configuration, see [Configure guardrails](https://www.infobip.com/docs/agentos-ai-agents/ai-agents/configure-guardrails). ___ ## Advanced settings Configure the prompt template for the knowledge agent. ### Prompt template [#prompt-template-advanced-settings] A custom instruction template that controls how context and the end user's question are formatted before being sent to the model. The default template is pre-configured and works for most use cases. Required variables (include without changes): - `{context_str}`: retrieved content from your knowledge sources - `{query_str}`: the end user's question Optional variable: - `{prompt_var}`: additional context passed at runtime NOTE The variables `{context_str}`, `{query_str}`, and `{prompt_var}` must be written in English. TIP Use delimiters such as `###` and `**` around variables to help the model distinguish between instructions and injected content. **Default template** ``` - Parts of the documentation: ###{context_str}### - Answer the users question: **{query_str}** ``` ___ ## Next steps Connect knowledge sources Give the agent content to retrieve from. Test the knowledge agent Test with sample questions before deploying. Use in a chatbot Add the knowledge agent to a chatbot or use it as a subagent within an orchestration. Manage knowledge agents Edit settings or delete agents. ___