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

# Knowledge Pack Optimization

> How to structure and prepare content so thinkrr agents deliver faster, more accurate answers when using Knowledge Packs.

## Overview

Knowledge Pack Optimization helps you structure content so thinkrr can deliver more accurate, faster responses with minimal latency. If you’re unfamiliar with how Knowledge Packs work or how to attach them, start with the [Knowledge Packs](/Knowledge-Packs) guide first.

## Best Practices

* Use `.md` (Markdown) or plain text whenever possible.
* Keep content relevant, concise, and well-structured.
* Write in short paragraphs or bullet lists to make concepts easier to process.
* Use clear, descriptive headings:
  * Start each topic with `##` headings.
  * Use `###` subheadings if a section grows too long.
* Group related information under the same section.
* Be specific: use names, dates, and numbers instead of “it” or “this.”
* Avoid images and tables — summarize visuals in text instead.
* Review formatting if agents provide inaccurate answers; clean Markdown works best.
* Use page-specific URLs instead of broad links.
* Keep only factual, customer-facing information. Do not include prompts or instructions.

## Recommended Formats

* **Preferred:** `.md` (Markdown) and `.txt`
* **Also supported:** `.pdf`, `.docx`, `.csv`, `.xlsx`\
  Keep documents text-based and avoid heavy formatting.

## Maintenance Tips

* Review and update packs whenever policies, pricing, or FAQs change.
* Remove outdated or conflicting content to prevent inaccurate answers.
* Track updates with a small “Changelog” section at the top if needed.

***

## FAQs & Troubleshooting

### General Questions

<Accordion title="Can I use one Knowledge Pack across multiple agents or projects?" icon="message-question" iconType="duotone">
  Yes. A single pack can be linked to multiple inbound agents and outbound projects.
</Accordion>

<Accordion title="Will larger Knowledge Packs slow responses?" icon="message-question" iconType="duotone">
  No. Thinkrr optimizes lookups, and latency is minimal (around 100ms). Large, text-heavy files are fine as long as they’re cleanly formatted.
</Accordion>

### Configuration

<Accordion title="What content should I avoid in a Knowledge Pack?" icon="gear" iconType="duotone">
  Do not include internal SOPs, credentials, or prompt text. Focus only on customer-facing facts, policies, and reference information.
</Accordion>

<Accordion title="Can sub-accounts access Knowledge Packs from an agency’s main account?" icon="gear" iconType="duotone">
  No. Sub-accounts cannot view or use Knowledge Packs created in an agency’s primary account.
</Accordion>

### Usage and Results

<Accordion title="What should I do if thinkrr gives an outdated or incorrect answer?" icon="book-open" iconType="duotone">
  Check that the Knowledge Pack content is current, remove outdated or duplicate details, and place the most important updates near the top of each section.
</Accordion>

<Accordion title="Should I include web URLs in Knowledge Packs?" icon="book-open" iconType="duotone">
  Yes, but only page-specific, public URLs. Avoid broad homepages or private/gated content that thinkrr cannot access.
</Accordion>

***

For additional questions or guidance, try using our [**Virtual Support Agent**](/Thinkrr-Support-Agent)! Available 24/7 to help resolve most issues quickly at [thinkrr.ai/support](https://thinkrr.ai/support).

If you still need assistance, visit our support site at [help.thinkrr.ai](https://help.thinkrr.ai) and submit a [Ticket](/Support-Ticket-System) or contact our team directly at [hello@thinkrr.ai](mailto:hello@thinkrr.ai).