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

# AI Agents

> AI across the partner lifecycle: a partner support agent, team copilot, submission approvals, deal coaching, and course and announcement authoring.

export const Headless = ({children, intro = true}) => {
  const [copied, setCopied] = useState(null);
  const copy = event => {
    const row = event.target.closest("li");
    if (!row) return;
    const text = row.textContent.trim();
    try {
      navigator.clipboard?.writeText(text);
    } catch {}
    setCopied(text);
    setTimeout(() => setCopied(null), 1600);
  };
  return <>
      {intro ? <p>
          Introw is <strong>headless</strong> by design: run everything here in plain language from Claude, ChatGPT,
          Slack, Teams, or your own CRM through Introw's <a href="/features/developer/mcp">MCP server</a>, with every
          action written straight back to your CRM source of truth. The portal is one surface, never the only one.
        </p> : null}

      {children ? <div className="not-prose" data-prompts="block">
          <div data-prompts="head">
            <span>{copied ? "Copied" : "Try it from your LLM"}</span>
          </div>
          <div data-prompts="rows" onClick={copy}>
            {children}
          </div>
          <a href="/headless/prompts" data-prompts="more">
            See all prompts
          </a>
        </div> : <Card title="See the headless motion" icon="wand-magic-sparkles" href="/headless">
          The agentic motion - lifecycle deep dives and a Claude Code skill library that run your partner program from
          Claude, ChatGPT, Slack, and your CRM.
        </Card>}
    </>;
};

export const Rail = ({children}) => <div className="not-prose" data-overview="rail">
    {children}
  </div>;

export const Impact = ({children}) => <div className="not-prose" data-overview="impact">
    {children}
  </div>;

export const Personas = ({children}) => <div className="not-prose" data-overview="personas">
    {children}
  </div>;

export const Pains = ({children}) => <div className="not-prose" data-overview="pains">
    {children}
  </div>;

> Introw is AI-native, not AI-bolted-on. Agents run across the entire partner lifecycle: a support agent answers partners on every channel, a copilot runs the program for your team, AI reviews and approves what partners submit, a coach works every deal, and AI authors your courses and announcements. Every agent is grounded in your CRM and a knowledge base you control, and the whole program is operable in plain language from the AI assistant your team already uses.

## The problem it solves

<Pains>
  | Without Introw | With Introw |
  | - | - |
  | Answers wait on your team | The agent answers, 24/7 |
  | Submissions queue for days | AI clears the routine ones |
  | You cannot coach every deal | A coach on every one |
  | Writing the content is the blocker | AI drafts it, you approve |
</Pains>

## Impact

Partners already work with AI. Being the vendor whose program answers inside it, coaches their deals and clears their requests in seconds is not a feature comparison, it is a different experience.

<Impact>
  for your business

  * **AI, not admin**
    Agents do the mechanical work across the lifecycle, with humans in the loop wherever judgment matters
  * **In your CRM**
    Agents ground answers, coaching and reviews in your CRM data and write outcomes back to the same records
  * **No new tool**
    Every agent meets people on the surface they use: the CRM, Slack, Teams, WhatsApp, email, or their assistant

  for your partners

  * **Self-serve**
    An answer in seconds, in their language, at any hour, without a person in the loop
  * **Enabled**
    A coach on their deal tells them what to do next, and which asset to use to do it
  * **Efficient**
    Their registration or claim is reviewed the moment it lands, not when somebody gets to it

  [A day in the life of your partners](/days-in-the-life)
</Impact>

<Personas>
  * **Partner Operations** - the agents and their knowledge
  * **Partner Marketing** - tone, training and announcements
  * **Sales Leadership** - coaching and conflict rules
  * **Partners** - answers and coaching, on demand
</Personas>

## How this area works

Introw runs a set of AI systems that take the mechanical work off your team, each grounded in your CRM and knowledge base and acting with humans in the loop. They fall into four groups:

* **Conversational agents** - a [partner support agent](./partner-support) that answers partners 24/7 on every channel, and an [AI copilot](./copilot) that lets your team run the program in plain language.
* **AI that reviews and guides** - [channel-conflict screening](./channel-conflict) and [AI approvals](./approvals) that validate what partners submit before it reaches a human, and a [deal coach](./deal-coaching) on every deal.
* **AI that creates** - [AI training](./training) that authors courses, tutors learners, and grades answers, and [AI announcements](./announcements) that draft partner comms.
* **The shared brain** - a [knowledge base](./knowledge-base) that grounds every one of them, extensible with custom context and live tools over MCP.

AI is woven through the rest of the platform too, where partners register and refer deals conversationally, intake arrives by email, and content is translated automatically.

Your CRM and knowledge base ground every agent, and the AI assistant is just another way in.

**Where this sits in a setup.** AI is not a phase. Every [setup track](/tracks) leaves it on from the start, because the support agent seeds itself from your own website the moment your workspace exists.

## Talk to Introw

Two conversational agents - one for your partners, one for your team:

<Rail>
  * ![The Partner Support Agent's configuration - its name, instructions, welcome message and quick-start prompts - beside the panel that tests it as a partner before it goes live.](https://assets.introw.io/docs/features/ai/partner-support/shots/agent.png)

    [**Partner Support Agent**](./partner-support)

    Answers partners 24/7, on every channel, and acts.

    [How to · 2 guides](./partner-support/technical)

  * ![Introw AI, the assistant for your own team, with one-click prompts for a QBR, inactive partners and gold partners and a box to ask anything about the program.](https://assets.introw.io/docs/features/ai/copilot/shots/assistant.png)

    [**AI Copilot**](./copilot)

    Run the program in plain language, and take action.

    [How to](./copilot/technical)
</Rail>

## AI that reviews and guides

AI checks what partners submit and coaches what your team works:

<Rail>
  * ![A form submission open for review, with the submitted fields on its activity thread and the partner it came from beside them.](https://assets.introw.io/docs/features/ai/approvals/shots/submission.png)

    [**AI Approvals**](./approvals)

    Clear the routine submissions, escalate the rest.

    [How to](./approvals/technical)

  * ![The Channel Conflict Analysis automation on a deal registration form, cross-referencing the CRM for an existing opportunity before the registration is accepted.](https://assets.introw.io/docs/features/ai/channel-conflict/shots/analysis.png)

    [**Channel Conflict**](./channel-conflict)

    Screen every registration against your pipeline.

    [How to · 1 guide](./channel-conflict/technical)

  * ![A deal coach's stage guidance, spelling out what must be achieved and how to achieve it at each pipeline stage, alongside objection handling and enablement.](https://assets.introw.io/docs/features/ai/deal-coaching/shots/coach.png)

    [**Deal Coaching**](./deal-coaching)

    A segment-specific coach on every deal.

    [How to · 1 guide](./deal-coaching/technical)
</Rail>

## AI that creates

AI does the content work your team used to do by hand:

<Rail>
  * ![The course editor with its AI assistant panel open, offering to rewrite the current chapter, add a quiz, or restructure the course.](https://assets.introw.io/docs/features/ai/training/shots/ai-course.png)

    [**AI Training**](./training)

    Author courses, tutor learners, grade open answers.

    [How to](./training/technical)

  * ![The AI announcement generator, asking what to announce and offering webinar, product update and blog post presets to draft from.](https://assets.introw.io/docs/features/ai/announcements/shots/ai-draft.png)

    [**AI Announcements**](./announcements)

    Draft from a link or a prompt, and get suggestions.

    [How to](./announcements/technical)
</Rail>

## The shared brain

<Rail>
  * ![The agent's knowledge sources, counting websites, snippets, documents, portal content and MCP servers, with the trusted domains it searches listed beside them.](https://assets.introw.io/docs/features/ai/knowledge-base/shots/sources.png)

    [**Knowledge Base**](./knowledge-base)

    The sources every agent answers from, kept live.

    [How to · 2 guides](./knowledge-base/technical)
</Rail>

## AI woven through the rest of Introw

Beyond the dedicated agents, AI does the drafting and data entry inside the rest of the platform. Partners register and refer conversationally. Intake arrives by email and is parsed for you. Company records autofill, and content is translated into every partner's language automatically.

<CardGroup cols={2}>
  <Card title="Register with AI" icon="file-signature" href="/features/deal-registration/registration/guides/ways-to-register">
    Resellers register deals conversationally, attributed in your CRM.
  </Card>

  <Card title="Refer with AI" icon="share-from-square" href="/features/referrals/lead-sharing/guides/ways-to-refer">
    Partners refer leads by email, chat, or an AI assistant, credited automatically.
  </Card>

  <Card title="Prep QBRs and reviews" icon="clipboard-list" href="./copilot">
    The copilot assembles a partner business review from live data.
  </Card>

  <Card title="Every language, automatically" icon="language" href="/features/localization">
    Courses, announcements, and forms translate into partners' languages.
  </Card>
</CardGroup>

## What happens to your data

Every AI feature in Introw runs through one gateway, and the data rules are enforced there rather than
in each feature. Every one of the agents above, every streamed answer, and every embedding behind
knowledge search routes only to model providers with **zero data retention** that also **do not train on
prompts or responses**. No feature can opt out of it by omission.

Practically, that means the partner data your agents reason over - CRM records, deal context, knowledge
base content, partner conversations - is used to answer the request and then not retained by the model
provider, and never becomes training data. It is the same principle as the rest of the platform: your
CRM stays the source of truth, and nothing about being [AI-native](/why/ai-native) requires handing your
data over. See [AI Security](./security) for the whole posture, in the form a security review asks for,
and [enterprise-grade](/why/enterprise-grade) for the wider governance picture.

Three questions that follow, answered plainly.
**Can you turn AI off?** Yes. The partner-facing agent does nothing until you enable it, and the
agents that assist your team are configured per feature, so a program can run with none of them on.
**Can you choose the model?** No. Model selection sits behind the gateway rather than in a
per-organisation setting, which is what lets the retention guarantee above be enforced centrally
instead of negotiated feature by feature.
**What happens when the agent cannot answer?** It says so and escalates to a human on your terms,
which you set in its instructions; a partner is never left with a confident wrong answer. See
[Partner support](./partner-support).

## Run it from your AI assistant

The whole program is operable in plain language from the AI assistant your team already uses. Introw runs a secure [MCP server](/features/developer/mcp), so Claude, ChatGPT, Gemini, or any MCP client can search partners, prep a QBR, check commissions and take scoped actions. They are the same actions the [AI copilot](./copilot) takes inside Introw. Partners get the same through [Partner Connect](/features/partner-connect) - their own assistant, scoped to their data. See the [headless vision](/headless) for the full picture.

<CardGroup cols={2}>
  <Card title="Connect your AI assistant" icon="plug" href="/features/developer/mcp">
    Run Introw from Claude, ChatGPT, Gemini, or any MCP client.
  </Card>

  <Card title="The headless vision" icon="wand-magic-sparkles" href="/headless">
    Run the whole program where work already happens.
  </Card>
</CardGroup>

<Headless>
  * Answer a partner's question about our MDF claim process.
  * Prep a QBR for Acme from live CRM data.
  * Review this MDF claim against our proof-of-spend rules.
</Headless>


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.