# https://inkeep.com/ llms-full.txt
# AI for Customer Experience.
Assistant for users
Copilot for support teams
Reports on content gaps
Get a demo
(on your own content)
Get a demo
(on your own content)
User
How do I add Inkeep to my docs?
Sparkles
AI Assistant
To add Inkeep to your documentation, follow these steps:
1. **1. Auto-magically onboard your sources [1](https://docs.inkeep.com/sources/onboard-new-sources)**
Add your docs, blogs, forums, and other content to create your knowledge base with our out-of-the-box connectors.
2. **2. Add a snippet of JavaScript [2](https://github.com/inkeep/starter-template-widgets-embed)**:
Add our search and chat components to any platform.
```javascript
Inkeep.ChatButton({
baseSettings: {
apiKey: "API_KEY",
primaryBrandColor: "#26D6FF"
}
});
```
4. **3. Monitor and improve your docs [3](https://docs.inkeep.com/overview/why-inkeep)**
Your team can view stats, user conversations, and our AI reports to identify documentation gaps and drive your content roadmap.
Sources
[Getting started](https://docs.inkeep.com/overview/why-inkeep)
[JS Widget QuickStart](https://github.com/inkeep/starter-template-widgets-embed)
+2
[](https://docs.anthropic.com/)[](https://www.postman.com/)[](https://docs.pinecone.io/home)[](https://docs.midjourney.com/)[](https://neon.tech/docs)[](https://posthog.com/questions)[](https://clerk.com/docs)[](https://solana.com/docs)
[](https://docs.anthropic.com/)[](https://www.postman.com/)[](https://docs.pinecone.io/home)[](https://neon.tech/docs)[](https://docs.midjourney.com/)[](https://solana.com/docs)
[](https://docs.anthropic.com/)[](https://www.postman.com/)[](https://docs.pinecone.io/home)[](https://neon.tech/docs)[](https://docs.midjourney.com/)[](https://solana.com/docs)[](https://clerk.com/docs)[](https://posthog.com/questions)[](https://render.com/docs)[](https://bun.sh/)
Trustedby leading enterprises and high-growth teams
## Assistant for self-serve support
* * *
Powered by your content
Help your users find what they need across all of your content.
Answers you can trust
Inkeep provides rich citations, admits when it doesn't know, and stays on topic.
Integrate anywhere
Add Inkeep to your docs, help center, app, and community in ~<10 minutes with our [quickstarts](https://docs.inkeep.com/overview/ai-for-customers#follow-a-quick-start).
## Copilot for your support team
Add an intelligent side panel to your ticketing platform. [watch demo](https://go.inkeep.com/keep-demo)
Draft Answers
Generate answers based on your previous tickets, Slack threads, and docs.
Summaries & To-dos
Get a summary of the conversation so far and actionable next steps.
Tickets into FAQs
Automatically turn resolved tickets into FAQs for your knowledge base.
Works with
ZendeskSalesforceHubspotIntercomHelpScout
## AI Agents for every step
* * *
Use Inkeep agents to automate any workflow and scale support.
Thoughtful auto-replies
Have your AI agent respond to emails or live chats β only when confident!
Intelligent support forms
Deflect tickets before they are created and simplify your support form to a single field.
Ticket routing & labeling
Categorize tickets, add custom labels, and route them to the correct team.
## Pinpoint and fix where users get stuck
* * *
Get actionable summaries on where there are gaps in your content.
Can I use as an internal Slack bot?
Integration
4
How often is content kept up to date?
Feature
3
How are forums used as a source?
Feature
5
Is there a plugin for Docusaurus?
Integration
8
How are LLM reports generated?
Feature
3
Can I use Inkeep with Zendesk?
Integration
12
## Ship with confidence
SOC 2 Type II compliance
No training on your data
Tools to monitor & tune
PII removal
Deflection reports
Auth & permissions
## Don't take our word for it
["Their platform has been instrumental in providing more efficient and effective self-serve support to our customers, and has also been fantastic for internal tooling. 1000% recommended!"\
\
\
\
Chris Coleman\
\
Head of Customer Success, Xano\
\
Try it](https://community.xano.com/)
["We saw much better out-of-the-box results compared to Algolia and it added a virtual team member that helps our users 24/7."\
\
\
\
Ruben Fiszel\
\
CEO, Windmill\
\
Try it](https://www.windmill.dev/)
["We have been really happy with Inkeep and how it powers our help docs and developer docs."\
\
\
\
JR Farr\
\
CEO, Lemon Squeezy\
\
Try it](https://docs.lemonsqueezy.com/help)
["User engagement is consistently high. Our content and product teams actively use the insights to improve our product and docs."\
\
\
\
Arnaud Comet\
\
Director of Product, SingleStore\
\
Try it](https://www.singlestore.com/sqrl)
["We have it on our Docs site and Discord server, and we're working on further integrations. The Inkeep team has been great!"\
\
\
\
Daniel Price\
\
Lead Technical Writer, Neon\
\
Try it](https://neon.tech/docs)
["We were wary at first, but the quality of answers sold us. My favorite feature: at the end of each week, we get an insight report that helps us prioritize product and docs improvements."\
\
\
\
Colin Sidoti\
\
CEO, Clerk\
\
Try it](https://clerk.com/docs)
["With Inkeep, we are able to significantly reduce our support volume with authentic self-serve experiences that match our brand."\
\
\
\
Zeno Rocha\
\
CEO, Resend\
\
Try it](https://resend.com/help)
["Our chatbot performance is significantly better since introducing Inkeep's tech, and our customers continue to provide us positive feedback on it. Definitely a product and team that understands what customers want. Inkeep's bots manage thousands of conversations each month for us allowing us to focus on higher quality human support."\
\
\
\
Lachlan McPherson\
\
Developer Advocate, Xano\
\
Try it](https://community.xano.com/)
["We evaluated the quality of the results are quite amazingly high, it can combine many sources and give a best answer based on the retrieved documents, our users also give us good feedback."\
\
\
\
Yi Wen\
\
Zilliz\
\
Try it](https://docs.zilliz.com/docs)
["Inkeep made it easy to add high quality AI support to our docs, blogs, and help center - reaching our users wherever they are."\
\
\
\
Bear Douglas\
\
Director of DevRel, Pinecone\
\
Try it](https://docs.pinecone.io/home)
["The quality is top-notch for both the search and chat. It required minimal effort from our team and our users are loving it."\
\
\
\
Alexander Klizhentas\
\
CTO, Teleport\
\
Try it](https://goteleport.com/chat/)
["Inkeep was stupidly easy to setup and implement in our Slack and Discord communities. Inkeep not only delights our customers with quick answers to queries on our public information, but it also takes the pressure off our support team by providing support to customers without the need to involve our support team. Plus, the Inkeep team are a delight to deal with! I would recommend Inkeep to anyone looking to AI powered search and support."\
\
\
\
Oli Wolff\
\
Product Manager, Kinde\
\
Try it](https://discord.gg/wHX6j7wG5d)
["The search was just plain better with zero config from us. Integrating into our docs was dead simple, truly seamless end-to-end."\
\
\
\
Nate Totten\
\
CTO, Zuplo\
\
Try it](https://zuplo.com/docs)
["Their platform has been instrumental in providing more efficient and effective self-serve support to our customers, and has also been fantastic for internal tooling. 1000% recommended!"\
\
\
\
Chris Coleman\
\
Head of Customer Success, Xano\
\
Try it](https://community.xano.com/)
["Inkeep made it easy to add high quality AI support to our docs, blogs, and help center - reaching our users wherever they are."\
\
\
\
Bear Douglas\
\
Director of DevRel, Pinecone\
\
Try it](https://docs.pinecone.io/home)
["With Inkeep, we are able to significantly reduce our support volume with authentic self-serve experiences that match our brand."\
\
\
\
Zeno Rocha\
\
CEO, Resend\
\
Try it](https://resend.com/help)
["We have been really happy with Inkeep and how it powers our help docs and developer docs."\
\
\
\
JR Farr\
\
CEO, Lemon Squeezy\
\
Try it](https://docs.lemonsqueezy.com/help)
["Inkeep was stupidly easy to setup and implement in our Slack and Discord communities. Inkeep not only delights our customers with quick answers to queries on our public information, but it also takes the pressure off our support team by providing support to customers without the need to involve our support team. Plus, the Inkeep team are a delight to deal with! I would recommend Inkeep to anyone looking to AI powered search and support."\
\
\
\
Oli Wolff\
\
Product Manager, Kinde\
\
Try it](https://discord.gg/wHX6j7wG5d)
["We evaluated the quality of the results are quite amazingly high, it can combine many sources and give a best answer based on the retrieved documents, our users also give us good feedback."\
\
\
\
Yi Wen\
\
Zilliz\
\
Try it](https://docs.zilliz.com/docs)
["We have it on our Docs site and Discord server, and we're working on further integrations. The Inkeep team has been great!"\
\
\
\
Daniel Price\
\
Lead Technical Writer, Neon\
\
Try it](https://neon.tech/docs)
["We were wary at first, but the quality of answers sold us. My favorite feature: at the end of each week, we get an insight report that helps us prioritize product and docs improvements."\
\
\
\
Colin Sidoti\
\
CEO, Clerk\
\
Try it](https://clerk.com/docs)
["We saw much better out-of-the-box results compared to Algolia and it added a virtual team member that helps our users 24/7."\
\
\
\
Ruben Fiszel\
\
CEO, Windmill\
\
Try it](https://www.windmill.dev/)
["The quality is top-notch for both the search and chat. It required minimal effort from our team and our users are loving it."\
\
\
\
Alexander Klizhentas\
\
CTO, Teleport\
\
Try it](https://goteleport.com/chat/)
["Our chatbot performance is significantly better since introducing Inkeep's tech, and our customers continue to provide us positive feedback on it. Definitely a product and team that understands what customers want. Inkeep's bots manage thousands of conversations each month for us allowing us to focus on higher quality human support."\
\
\
\
Lachlan McPherson\
\
Developer Advocate, Xano\
\
Try it](https://community.xano.com/)
["User engagement is consistently high. Our content and product teams actively use the insights to improve our product and docs."\
\
\
\
Arnaud Comet\
\
Director of Product, SingleStore\
\
Try it](https://www.singlestore.com/sqrl)
["The search was just plain better with zero config from us. Integrating into our docs was dead simple, truly seamless end-to-end."\
\
\
\
Nate Totten\
\
CTO, Zuplo\
\
Try it](https://zuplo.com/docs)
["Their platform has been instrumental in providing more efficient and effective self-serve support to our customers, and has also been fantastic for internal tooling. 1000% recommended!"\
\
\
\
Chris Coleman\
\
Head of Customer Success, Xano\
\
Try it](https://community.xano.com/)
["We were wary at first, but the quality of answers sold us. My favorite feature: at the end of each week, we get an insight report that helps us prioritize product and docs improvements."\
\
\
\
Colin Sidoti\
\
CEO, Clerk\
\
Try it](https://clerk.com/docs)
["Inkeep made it easy to add high quality AI support to our docs, blogs, and help center - reaching our users wherever they are."\
\
\
\
Bear Douglas\
\
Director of DevRel, Pinecone\
\
Try it](https://docs.pinecone.io/home)
["We saw much better out-of-the-box results compared to Algolia and it added a virtual team member that helps our users 24/7."\
\
\
\
Ruben Fiszel\
\
CEO, Windmill\
\
Try it](https://www.windmill.dev/)
Level up your self-serve and support experiences today
Get a demo
(on your own content)
Get a demo
(on your own content)
## Latest Posts
* * *
[\
\
**Inkeep MCP & RAG API**\
\
Power MCP Servers and AI Agents with your own data.\
\
Nick Gomez\
\
March 25, 2025](https://inkeep.com/blog/mcp-server)
[\
\
Product Launch\
\
**Keep - A Copilot for Support Teams**\
\
An AI sidekick designed for high-touch human support.\
\
Nick Gomez\
\
November 1, 2024](https://inkeep.com/blog/copilot-for-support-teams)
[\
\
Launch week\
\
**OpenAI-compatible APIs**\
\
Build your own custom experiences with our OpenAI-compatible chat and analytics APIs.\
\
Nick Gomez\
\
August 9, 2024](https://inkeep.com/blog/use-your-own-ui)
[\
\
Launch week\
\
**Welcoming Andrew & Omar**\
\
Introducing Andrew Mikofalvy and Omar Gonzalez to the Inkeep team!\
\
Nick Gomez\
\
August 8, 2024](https://inkeep.com/blog/welcoming-andrew-mikofalvy-and-omar-gonzalez)
[\
\
Launch week\
\
**In-App Copilots & Workflows**\
\
Create personalized experiences that adapt to your user's journey.\
\
Nick Gomez\
\
August 7, 2024](https://inkeep.com/blog/in-app-copilots)
[\
\
Launch week\
\
**UI Kit: One hub for all user questions**\
\
Embed AI-powered search, chat, and support as one experience, anywhere.\
\
Nick Gomez\
\
August 6, 2024](https://inkeep.com/blog/uikit)
[\
\
Launch week\
\
**Slack & Discord: Support Copilots**\
\
Use our 1-click install bots to serve your community or support team.\
\
Nick Gomez\
\
August 5, 2024](https://inkeep.com/blog/slack-and-discord-copilots)
[Back](https://inkeep.com/blog)
Launch week
August 5, 2024
# Slack & Discord: Support Copilots
Nick Gomez
Whether it's Discord for your community or 1:1 Slack channels with enterprise customers, many modern teams operate and engage with customers on these platforms.
Today, we're excited to share new features that help supercharge these workflows.
## 1. The `Ask for help π` button
Sometimes, customer-facing AI needs a human touch.
When enabled, users can click on the **`Ask for help π`** button included in every AI message to tag your team for additional assistance. The bot can tag users, groups, and even other bots for more advanced workflows.
With this button, teams can tap into AI support while maintaining a culture of hands-on support.
## 2. `Auto-Reply` mode for your community
The Inkeep Discord bot now supports Auto-Reply functionality in forum-styled threads in addition to traditional channels. Forum-styled threads provide a more structured way to keep user conversations in discrete threads. With Auto-Reply functionality, it's more streamlined than ever to add an `β¨ask-ai` channel to your community.
## 3. Modes designed for support teams
Have `#support-triage` or `#ask-an-expert` channels where your team coordinates customer inquiries?
We've made a few tools designed for you:
- **`Auto-Reply` & `AI Draft` modes**: When both modes are enabled, Inkeep will chime in on every new thread of select channels with helpful, human-like responses your team can use as quick initial drafts to user questions.
- **`Loop-In` mode**: Alternatively, tag the Inkeep bot only when relevant. It'll naturally jump in and provide context-aware suggestions based on the conversation so far.
Whether you prefer automatic, always-there suggestions or to selectively loop in the Slack bot, Inkeep plugs in to whichever way your team wants to work.
## 4. Custom branding
**Special bonus**: you can now customize the Inkeep bot to match your brand with a custom logo (for Slack) and name (for Slack & Discord).
## Getting started
Check out the [Slack](https://docs.inkeep.com/integrations/slack) or [Discord](https://docs.inkeep.com/integrations/discord) integration documentation.
Follow Inkeep product updates
Email
Subscribe
[Back](https://inkeep.com/blog)
Launch week
August 7, 2024
# In-App Copilots & Workflows
Nick Gomez
Today, we're happy to share 3 powerful features of Inkeep's UI Kit that let teams easily create hyperpersonalized in-app experiences for their users.
Let's dive in by using Inkeep's Copilot as an example.
## 1. Context: Ground the model with user information
First, let's use **Context** to pass in key information about the user or the current state of our application.
```typescript
const aiChatSettings: InkeepAIChatSettings = {
//... other aiChatSettings
context: `[
"The user is currently viewing the app page ${currentUrl}.",
"The user has access to the following features: ${featureFlags.join(', ')}",
"The user is on a ${user.planType} plan.",
"The user is using the ${productVersion} version of the application."
]`
};
```
This information helps the AI assistant better understand user questions and provide tailored answers specific to their situation.
## 2. Guidance: Steer the conversation
**Guidance** can be used to customize tone, style, and provide instructions for how to interact with the user.
These can be set globally on the project settings in the Inkeep dashboard, but we can also pass in additional or runtime guidance as part of our UI Kit component.
```typescript
// if user.createdAt < 7 days
const aiChatSettings: InkeepAIChatSettings = {
//... other aiChatSettings
guidance:`
Clarify documentation type
When the user mentions adding Inkeep to their documentation
And it is unclear what documentation platform they are using
Briefly list common documentation platforms
and ask which one they're using.
Suggest exploring features
After answering a user's initial question
Briefly mention the Slack or Discord bots as integration options.
Offer additional support
When there's ambiguity in the user's understanding
Or when the user seems to need extra assistance
Remind the user that the Inkeep team is always ready to help.
`
};
```
We can experiment with how different combinations [impact our user funnels](https://docs.inkeep.com/customization-guides/use-your-own-analytics) for different segments of users.
## 3. Workflows: Guide users through common tasks
Workflows allow you to define goal-oriented scenarios like onboarding tasks or troubleshooting sequences.
When a user clicks on a workflow button at the start of a chat, the AI assistant will focus on helping the user with that particular scenario.
For an example, try the **Help me integrate with my app** workflow on our [landing page](https://inkeep.com/).
Here's what the underlying configuration looks like:
```typescript
const integrateWithMyAppWorkflow: Workflow = {
id: "ikp_integrate_with_my_app",
displayName: "Help me integrate with my app",
goals: [ "Identify the platform that the user is looking to add Inkeep on.", "Give instructions for how to add the bot to that platform." ],
botPersona: "You are an expert solutions engineer",
informationToCollect: [ { description: "Platform they are integrating Inkeep to", required: true, }, ],
guidance: [ "Ask clarifying questions to understand the underlying platform.", "Provide platform-specific integration instructions.", ],
initialReplyMessage:
"Happy to help you integrate Inkeep. Where are you looking to add it?",
};
const aiChatSettings: InkeepAIChatSettings = {
// ... other aiChatSettings
workflows: [integrateWithMyAppWorkflow]
};
```
## Bringing it all together
Here's a high level strategy we might combine context, guidance, and workflows for two different user types at Inkeep:
### _New User_: Reaching the "Aha" Moment
1. **Context:** Provide information from the acquisition funnel about what the user might be interested in.
2. **Guidance:** Emphasize quick wins and maintain a friendly, encouraging tone.
3. **Workflows:** Offer a "Quick Start" guide for key setup steps.
### _Experienced User_: Getting closer to enterprise features
1. **Context:** Provide a list of core activation tasks the user has already performed.
2. **Guidance:** Focus on advanced customizations and technical depth/support.
3. **Workflows:** Provide a "Learn about X" workflow related to an enterprise feature
These configurations can be easily managed and changed as part of our application state and passed to Inkeep's UI Kit React components. Get started building your own [In-App Copilot](https://docs.inkeep.com/customization-guides/workflows) today.
Follow Inkeep product updates
Email
Subscribe
[Back](https://inkeep.com/blog)
Launch week
August 6, 2024
# UI Kit: One hub for all user questions
Nick Gomez
Whether it's searching docs, asking in a community forum, or reaching out to support, users have one goal: **getting answers to their questions**.
Today, we're excited to announce `@inkeep/uikit`, our refreshed library of out-of-box search, chat, and support UI components. You can easily embed these components as 1st class React components or simple JavaScript snippets.
## 1. Upgraded Search Experience
Our new single-pane design brings a fresh take on what modern search should look like.
Beyond a new design, also find:
- Customizable tabs if you have multiple sources
- Better keyboard navigation
- A toggle to explicitly switch between `Search` and `Ask AI` modes
- An optional `Ask AI` result card that triggers the AI assistant with the current search query
## 2. Refreshed Chat UI
Our chat UI features redesigned citation cards that elegantly show key information at a glance: source type, breadcrumbs, and page title. Along with the inline footnotes ( [1](https://footnote.com/)) in the answer, users can quickly understand where answers are coming from and go learn more straight from the source.
## 3. Embedded Support Escalation
Our new [embedded support form](https://docs.inkeep.com/customization-guides/embedded-support-form) makes human-help easily accessible right from the search or AI chat experience. Users can access it via the `Get help` button, which can also link to your community or other support channels, or through the `Contact us` button that intelligently appears when the AI assistant can't find an answer.
The embedded form is fully customizable and connects to any downstream support platform like Zendesk or HelpScout.
## Getting Started
UI Kit offers a unified solution for search, chat, and support β all in one beautiful and easy-to-embed experience.
Use one of our [15+ quickstarts](https://docs.inkeep.com/overview/ai-for-customers#follow-a-quick-start) to embed UI Kit to your docs, landing page, or app today.
Follow Inkeep product updates
Email
Subscribe
[Back](https://inkeep.com/blog)
March 25, 2025
# Inkeep MCP & RAG API
Nick Gomez
This week we launched a new API: `inkeep-rag`, which lets you query your docs, help center content, knowledge bases, and any other source ingested by Inkeep.
With this API, you can use your own content to power custom support agents, copilots, or custom workflows.
Let's see how to use it as part of an MCP Server, so we can connect it to Cursor or Anthropic's Claude Desktop App.
## API Request
The `inkeep-rag` API takes in a query and outputs documents or chunks of content that can be passed directly to an LLM.
The API is exposed in the OpenAI chat completions format. The βqueryβ is actually just a messages array, making it interoperable with the formats used by OpenAI, Anthropic, or other LLM providers. We can use any OpenAI compatible SDK to make the request.
```python
from openai import OpenAI
client = OpenAI(
base_url=INKEEP_API_BASE_URL,
api_key=INKEEP_API_KEY,
)
response = client.chat.completions.create(
model="inkeep-rag",
messages=[ {"role": "user", "content": "How do I get started with Inkeep?"}, ]
)
```
The `messages` array can contain full conversations, system prompts, and even images. The Inkeep RAG service automatically handles any summarization or intermediary steps for finding the most relevant content for the conversation.
## API Response
The `inkeep-rag` API responds with `content` containing an array of documents, each in the format of:
```jsx
{
"type": "document",
"source": {
"type": "text",
"media_type": "text/plain",
"data": "To get started with Inkeep..."
},
"title": "Getting Started",
"context": "This is a docs page.",
"record_type": "documentation", // inkeep specific field
"url": "https://docs.inkeep.com/overview/ai-for-customers", // inkeep specific field
}
```
## Using the content
This format extends Anthropic's [Claude Citations](https://docs.anthropic.com/en/docs/build-with-claude/citations), so can easily be used by Anthropic's Claude models and generally by any LLM or tool.
To call Claude directly:
```python
import anthropic
client = anthropic.Anthropic()
# query = "How to get started with Inkeep?"
documents_from_inkeep_rag_model = [...]
response = client.messages.create(
model="claude-3-7-sonnet-20250219",
max_tokens=1024,
messages=[ { "role": "user", "content": [ *documents_from_inkeep_rag_model, { "type": "text", "text": "Write a blog post on how to get started" } ] } ]
)
```
## Using it within an MCP Server
To use this RAG model with Cursor, Anthropic's Claude Desktop App, or other MCP compatible clients, use [this](https://github.com/inkeep/mcp-server-python/blob/main/inkeep_mcp_server/server.py) template to get started.
You can customize the MCP Server with any business logic you like. As a basic example, we can create an MCP Tool that takes in a `query` and calls the Inkeep RAG API.
```python
import openai
from mcp.server.fastmcp import FastMCP
from pydantic import BaseModel
mcp = FastMCP("inkeep-mcp-server")
@mcp.tool(
name="retrieve-product-docs",
description="Retrieve product docs about your product using the Inkeep RAG API. 'query' should be framed as a conversational question about the product.",
)
async def retrieve_product_docs(query: str) -> InkeepRAGResponse:
# make the call to Inkeep's RAG API
return res
```
## Getting started
As you can see - we lean on using formats and standards already common and established so it's easy to plug in Inkeep to any LLM application or use case.
To get started with our new RAG API, check out the [open source repo](https://github.com/inkeep/mcp-server-python) to set up your own local MCP server powered by your own content.
Follow Inkeep product updates
Email
Subscribe
[Back](https://inkeep.com/blog)
Product Launch
November 1, 2024
# Keep - A Copilot for Support Teams
Nick Gomez
Inkeep - Copilot for Support Teams from Nick Gomez on Vimeo
Playing in picture-in-picture
Share
Play
00:00
03:14
Settings
QualityAuto
SpeedNormal
Fullscreen
We know many teams care deeply about providing excellent human-based support.
That's why we're excited to announce **Keep**, a copilot designed to help your support team provide quick, high-quality answers without taking them out of the loop.
Here's how it works.
## Smart Assist
When you open Keep, it:
1. Understands the current state of the ticket
2. Inspects both your public & private content
3. Generates intelligent suggestions to move the ticket forward
### Draft Answers
If Keep feels confident that it can provide a high-quality answer that likely resolves the ticket, it'll generate a **draft answer** for you.
You can use the quick actions to make it shorter, more concise, or simply follow-up with a description of how to modify the answer. **Keep is fully conversational!**
Once you're happy with it, click on **Apply** to prefill it into the input box. There, add any final touches and then submit to the user.
### Quick Links
Keep will always show you an overview of the most relevant sources for the ticket.
Keep uses your public sources, like your docs, but can also leverage private sources like **previous support tickets** and **Slack conversations**.
### Summaries & To-Do's
For longer conversations, Keep will generate a **summary** of the conversation so far.
..as well as provide glanceable hints on the remaining **to-dos**:
### Other hints
Keep can also generate **quick replies**, call out when the user has expressed **strong emotions**, and provide other helpful suggestions.
This is all _automatic_; Keep is tuned to provide what's most relevant without adding noise.
## Close the content loop
With Keep, you can turn closed tickets into publishable FAQs with one click:
You can add the FAQ to your Inkeep knowledge base, copy it as markdown for your docs, or download it as a file.
## Getting Started
You can use Keep on Zendesk as a sidebar app or on any support platform via the Chrome Extension.
Want to see how it works with your content? [Schedule a demo](https://go.inkeep.com/demo).
## Extras
### How it's built
For those curious, here's some key technical details:
- The experience is powered by our `inkeep-qa` and `inkeep-context` [AI APIs](https://go.inkeep.com/ai-api). They're powered by the Inkeep RAG service and use [Claude Sonnet 3.5](https://docs.anthropic.com/en/docs/about-claude/models) as the output LLM model.
- Since our APIs follow the OpenAI chat-completions format, they can be used with any LLM app framework or tooling.
- This includes the [Vercel AI SDK](https://sdk.vercel.ai/), which we use to optimistically stream React components as Keep chooses the most relevant ones for a given support ticket.
- Check out our [quickstarts](https://github.com/inkeep/inkeep-chat-api-examples) for using the Inkeep AI APIs.
### Inspiration
We think chat UIs that leverage interactive, visually compelling UI components over just plain-text make the experience of digesting and iterating with AI copilots so much better.
Great examples include:
- [V0](https://v0.dev/) - Copilot for creating dynamic UI components.
- [Cursor](https://www.cursor.com/) - AI IDE for software development.
- [Claude Artifacts](https://support.anthropic.com/en/articles/9487310-what-are-artifacts-and-how-do-i-use-them) - reusable, referenceable "artifacts" for a general purpose AI assistant.
Our goal was to create a similarly immersive and interactive experience for support teams.
Follow Inkeep product updates
Email
Subscribe
[Back](https://inkeep.com/blog)
Launch week
August 8, 2024
# Welcoming Andrew & Omar
Nick Gomez
We are thrilled to announce two new additions to our growing team at Inkeep: [Andrew Mikofalvy](https://www.linkedin.com/in/andrew-mikofalvy/) and [Omar Gonzalez](https://www.linkedin.com/in/omar-gonzalez-200a35b4/)!
**Andrew Mikofalvy - Founding Engineer**
Andrew joins us as a Founding Engineer, bringing a wealth of experience in AI. He graduated from MIT with a degree in Aerospace Engineering and previously led AI products as Director of Engineering at Rev.
**Omar Gonzalez - Founding Engineer**
Omar also joins us as a Founding Engineer. He holds a degree in Computer Science from MIT and was formerly a Software Engineer at Meta, where he specialized in developing full-stack web applications.
**Elevating Our Team**
Andrew and Omar's arrival marks a significant milestone for Inkeep, we are so excited to have them onboard. Their expertise will help us build the next generation of AI-powered search and support solutions. Stay tuned for exciting updates as Andrew and Omar bring their talents to Inkeep and help shape the future of our company!
Follow Inkeep product updates
Email
Subscribe
[Back](https://inkeep.com/blog)
Launch week
August 9, 2024
# OpenAI-compatible APIs
Nick Gomez
Today, we're excited to share our new AI Chat and Analytics APIs that make it easier than ever to create your own custom support or conversational experiences with Inkeep.
The APIs are designed to follow the OpenAI chat completions-API format. This allows you to use existing tools for developing LLM applications while getting all of the benefits of Inkeep's service: managed content ingestion, grounded answers, and analytics for your team.
## AI Chat API
Here's a quick example using the OpenAI JavaScript SDK with Inkeep's AI Chat API. We just need to change the baseUrl and use `inkeep-qa-expert` as the model.
```javascript
import OpenAI from 'openai';
const client = new OpenAI({
baseURL: 'https://api.inkeep.com/v1/',
apiKey: process.env.INKEEP_API_KEY,
});
// initial chat
let chat = {
messages: [ { role: "user", content: "How do I get started with Inkeep?" } ],
};
const res = await client.chat.completions.create({
model: 'inkeep-qa-expert',
messages: chat.messages,
});
// add to convo history
chat.messages.push({
role: 'assistant',
content: res.choices[0].message.content,
});
```
Optionally, you can register a `provideLinks` tool to get back a list of sources that were used to generate the answer. Check out [this](https://github.com/inkeep/inkeep-chat-api-examples/blob/main/examples/qa-mode/openai-node-sdk-qa/index.ts) repo for a full example.
## Analytics API
### **Log a Conversation**
Our new Analytics API enables you to log any OpenAI-compatible conversation to Inkeep's Analytics service.
Let's use it on the results of the previous conversation:
```javascript
const loggedConv = await fetch('https://api.analytics.inkeep.com/conversations', {
method: 'POST',
headers: {
'Authorization': 'Bearer YOUR_API_KEY',
'Content-Type': 'application/json'
},
body: JSON.stringify({
type: "openai",
...messages: chat.messages
})
});
```
Any conversation logged to this endpoint benefits from Inkeep's rich dashboard, analytics, and reporting features.
### **Submit Feedback**
Now let's use the `/feedback` endpoint to log end-user feedback:
```javascript
const feedbackData = {
type: "positive",
messageId: loggedConv.messages.at(-1).id,
reasons: [{ label: "accurate_code_snippet", details: "The code worked perfectly." }]
};
fetch('https://api.analytics.inkeep.com/feedback', {
method: 'POST',
headers: {
'Authorization': 'Bearer YOUR_API_KEY',
'Content-Type': 'application/json'
},
body: JSON.stringify(feedbackData)
});
```
With this, you can create and monitor your own feedback mechanism like π/π functionality.
### Other primitives
The Inkeep Analytics API also provides all the primitives you need to create and monitor your LLM chat application:
- An `/events` endpoint to log analytics for custom user interactions like `answer_copied` or `chat_shared`.
- A `GET /conversations` endpoint to fetch existing conversations. You can leverage this to add "Chat history" or "Share chat" functionality to your chat experience.
## Get Started
Build your own chat experience today by leveraging the [Chat](https://docs.inkeep.com/ai-api/chat-completions-api) and [Analytics](https://docs.inkeep.com/analytics-api/add-analytics-to-any-chat) APIs.
Follow Inkeep product updates
Email
Subscribe
AI for CustomersAI for CX Teams
## Answer user questions, anywhere.
Showcase
'Ask AI' assistant
Used byAnthropic
Unified search
Used byPinecone
Help center
Used byLemon Squeezy
Slack bot
Used byArangoDB
Discord bot
Used byNeon
In-product copilot
Used bySingleStore
Custom copilot
Used byPostman
'Ask AI' assistant
[See it live](https://docs.anthropic.com/en/docs/welcome)
Use a [Chat Button](https://docs.inkeep.com/ui-components/js-snippet/chat-button) to add an **Ask AI** experience to your marketing site, docs, or app.
## All the tools your team needs to scale support.
Showcase
Support copilot
Copilot for support teams
Slack assist
Smart form
Intelligent support form
Auto-reply for live chat
Email Reply
Auto-reply for email
Report
Content gap reports
Tickets to FAQs
Support copilot
Copilot for support teams
[Learn more](https://docs.inkeep.com/overview/ai-for-support-teams)
Use Keep, our copilot for support teams, as a sidekick for providing [fast human-in-loop support](https://inkeep.com/blog/copilot-for-support-teams). Keep drafts answers, summarizes threads, and calls out to-dos. Install as a sidebar within your existing support platform.
Inkeep - Copilot for Support Teams from Nick Gomez on Vimeo
Playing in picture-in-picture
Share
Play
00:00
03:14
Settings
QualityAuto
SpeedNormal
Fullscreen
Ready to get started?
[Try our quickstarts](https://docs.inkeep.com/overview/ai-for-customers#follow-a-quick-start).
## Meet the team
* * *
We're a team of engineers passionate about machine learning, data engineering, and delightful user experiences.
[\
\
Nick Gomez\
\
Founder, CEO\
\
Nick studied Business Analytics & Computer Science at MIT. He helped lead devex for Microsoft's identity products.](https://www.linkedin.com/in/edwinngomez/)
[\
\
Robert Tran\
\
Founder, CTO\
\
Robert has a Master's in Computer Science from MIT. He architected search and data pipelines as Head of Engineering at illumis (acquired).](https://www.linkedin.com/in/robert-tran-profile/)
[\
\
Tim Cardona\
\
Founding ML Engineer\
\
Tim has a degree in Physics and Computer Science from MIT and worked at BlackRock as an ML and NLP Researcher.](https://www.linkedin.com/in/timothy-cardona/)
[\
\
Andrew Mikofalvy\
\
Founding Engineer\
\
Andrew graduated from MIT with a degree in Aerospace Engineering and led AI products as Director of Engineering at Rev.](https://www.linkedin.com/in/andrew-mikofalvy/)
[\
\
Sarah Niemiec\
\
Founding Design Engineer\
\
Sarah has a degree in Art & Design from the Milwaukee Institute of Art and is an experienced front-end engineer and designer.](https://www.linkedin.com/in/sarah-niemiec/)
[\
\
Omar Gonzalez\
\
Founding Engineer\
\
Omar has a degree in Computer Science from MIT and was a Software Engineer at Meta developing full-stack JavaScript applications.](https://www.linkedin.com/in/omar-gonzalez-200a35b4/)
[\
\
Miles Kaming-Thanassi\
\
Founding Engineer\
\
Miles studied Electrical Engineering and Computer Science at MIT and has a background in search algorithms.](https://www.linkedin.com/in/miles-kaming-thanassi-00307b194/)
[\
\
Justin Chavez\
\
Founding Engineer\
\
Justin studied CS and neuroscience at UPenn and UMBC. He has been working with AI startups since 2018 as a software and ML engineer.](https://www.linkedin.com/in/justinlchavez/)
[](https://linkedin.com/in/anubra266)
[Abraham Aremu\
\
Founding Engineer](https://linkedin.com/in/anubra266)
[Abraham is the creator of](https://linkedin.com/in/anubra266) [Choc UI](https://choc-ui.com/) and core maintainer of [Chakra UI](https://chakra-ui.com/), building accessible component libraries for modern web applications.