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

# Lead sources: add, edit limits, pause

> Give an agent something to work on. Add a source, change its limits, and read the state it reports back.

export const UiPath = ({children}) => {
  const parts = String(children).split(/\s*(?:→|>)\s*/).filter(Boolean);
  return <span className="pl-path">
      {parts.map((part, index) => <span key={`${index}-${part}`}>
          {index > 0 && <span className="pl-path__sep">→</span>}
          {part}
        </span>)}
    </span>;
};

export const Screenshot = ({id, alt, caption, frame = 'browser', url, marks = [], lang = 'en', workspace = 'northwind-outbound'}) => {
  const src = `/images/screenshots/${lang}/${id}.png`;
  const chrome = url ? `app.pipelime.ai/${workspace}${url}` : 'app.pipelime.ai';
  return <figure className={`pl-shot pl-shot--${frame} not-prose`}>
      <div className="pl-shot__frame">
        {frame !== 'bare' && <div className="pl-shot__bar">
            <span className="pl-shot__dots">
              <span className="pl-shot__dot" />
              <span className="pl-shot__dot" />
              <span className="pl-shot__dot" />
            </span>
            <span className="pl-shot__url">{chrome}</span>
          </div>}
        <div className="pl-shot__media">
          <img src={src} alt={alt} loading="lazy" />
          {marks.map(mark => <span key={mark.n} className="pl-shot__mark" style={{
    left: `${mark.x}%`,
    top: `${mark.y}%`
  }} aria-hidden="true">
              {mark.n}
            </span>)}
        </div>
      </div>
      {caption && <figcaption className="pl-shot__caption">{caption}</figcaption>}
    </figure>;
};

By the end of this page an agent has a lead source feeding it: sized and paced the way you want, and pausing or resuming with one click.

<div className="pl-availability">
  <div className="pl-availability__row">
    <div className="pl-availability__label">Where</div>
    <div className="pl-availability__value"><span className="pl-path">Outreach<span className="pl-path__sep">→</span>AI Agents</span></div>
  </div>

  <div className="pl-availability__row">
    <div className="pl-availability__label">Your role needs</div>
    <div className="pl-availability__value">Read access to AI Agents <code>read-campaigns</code>. Admin, Member and Viewer have it by default.</div>
  </div>

  <div className="pl-availability__row">
    <div className="pl-availability__label">To create or change</div>
    <div className="pl-availability__value"><code>create-campaigns</code> to add one, <code>update-campaigns</code> to change one, on top of the permission above.</div>
  </div>

  <div className="pl-availability__note">If you cannot find this in your sidebar, your workspace may have a custom menu configuration. Contact support and we will check it for you.</div>
</div>

## Before you begin

* An agent to attach the source to. The Sources tab lives inside an agent's own page, and a source you add there is joined to the agent through the agent's tag, created for you if the agent doesn't have one yet.
* A role that can change sources, which is gated separately from reading the agent. **Add a source** needs permission to create sources; the limit pencils, the pause button and **Edit source** need permission to update them; **Delete source** needs permission to delete them. A read-only role sees every source and every status, with those controls hidden rather than greyed out.

## Steps

<Steps>
  <Step title="Open the Sources tab">
    Go to <UiPath>Outreach → AI Agents</UiPath>, open an agent, and choose the **Sources** tab. Each source
    is a saved search, shown as one row: its name and status chip on the left; three stats in the middle:
    **Companies** (how many it has analyzed, against the total you set), **Daily** (its per-day cap) and
    **Leads** (how many people it has found so far, which loads a moment after the row); and on the right
    its pause button plus a menu: **View leads**, **View companies**, **Delete source**, and **Edit source**
    on any source built with the AI Engine search, which is every source added from this tab. Filter chips
    (**All**, **Active**, **Paused**, **Finished**) and a search box narrow a long list, and **Refresh**
    re-checks it without leaving the tab.

    <Screenshot id="ai-agents/lead-sources--sources-tab" url="/ai-agents/12?tab=sources" alt="The Sources tab showing one source's row: its name, an Active status chip, and the Companies, Daily and Leads stat columns" caption="Every source an agent works from lives on this one tab, whatever state it's in." />
  </Step>

  <Step title="Add a source">
    Choose **Add a source** to open the dialog: the same dialog also opens on its own if you reach it
    through **Add leads**, then **Lead generation**, instead. Tell it what to search for: **Industry** terms,
    **Company locations**, **Decision-maker roles**, or expand **Advanced targeting** for **Company size**
    and **Exclude locations**. **Suggest with AI** drafts the industry terms, roles, company sizes and
    locations from your workspace's company profile, and aims at something the sources you already have
    don't cover. **Source name** is the only field the dialog insists on beyond a filter, and it fills
    itself in from your first industry terms and location until you type your own.

    Turning on **Also find phone numbers** is worth pausing on: it's what puts the **Phones** badge on the
    source's row afterwards, because finding a number [costs more than finding just an email](/en/finding-leads/credits-per-lead).
    Expand **Pace & limits** to set **Total companies**, **Companies per day** (capped by your plan) and
    **Contacts per company**. **Quality gate (optional)** skips companies whose AI match score falls short
    of a bar you set. Choose **Create source** to save. The source is joined to the agent through the
    agent's tag (created first if the agent didn't have one), and sourcing is started for you straight
    away.

    <Screenshot id="ai-agents/lead-sources--add-source" url="/ai-agents/12?tab=sources" alt="The Add a source dialog with an industry term entered as a chip and a source name typed in, the Pace & limits section still collapsed" caption="Only a filter and a name are required: everything under Pace & limits has a working default." />
  </Step>

  <Step title="Change a source's limits without opening the dialog">
    Back on the Sources tab, the pencil beside **Companies** or **Daily** opens a small popover (a
    stepper, a slider, and **Save**), so you can raise or lower either cap in a couple of clicks, no need
    for the full dialog. On a source whose status reads **Max Companies** or **Finished**, the same popover
    is one click away behind an **Increase limit** link next to the pause button; raising the number
    restarts a **Max Companies** source, but not a **Finished** one.

    <Screenshot id="ai-agents/lead-sources--quick-edit-limits" url="/ai-agents/12?tab=sources" alt="The quick-edit popover open on a source's Total companies stat, showing a stepper, a slider and a Save button" caption="The same popover opens from a source's pencil at any time, or from Increase limit once a source has reached its company limit." />
  </Step>

  <Step title="Pause or resume one source">
    The play/pause button on a source's row touches that source alone: pausing it stops it from finding
    new companies, while the agent's outreach and every other source keep running untouched. Only a source
    you paused yourself, or one still sitting at **Pending**, comes back with the same button; a source
    that stopped for any other reason needs that cause fixed first, and the button stays disabled until it
    is (more on that below).
  </Step>
</Steps>

## What happens next

<Check>
  The source is on the Sources tab carrying a **Pending** or **Active** chip, and its **Companies** count
  starts climbing toward the total you set.
</Check>

A brand-new source sits at **Pending** until the searches behind it are built. The agent's page keeps
re-checking in the background for a while after that, so it flips to **Active** on its own with no manual
refresh. If it stays at **Pending**, press play on its row: the button that resumes a paused source starts
a pending one too. If a source you're not watching ever stops on its own, its status chip says why, and
the next section covers what each reason needs.

One more thing worth knowing: the tag is what joins a source to an agent, not a stored link to that one
agent. So if no agent's tag matches a source any more (the agent was deleted, or its tag changed), the
source isn't deleted with it. It turns up on the AI Agents list as a lead search not connected to an
agent, with **Finish setup** to build a new agent around it.

## Troubleshooting

<AccordionGroup>
  <Accordion title="The play button on a source's row is greyed out">
    Three statuses disable it, and each needs a different fix: none of them the pause button.

    **Max Companies**: the source has analyzed as many companies as its **Total companies** limit allows.
    Raise that number from **Increase limit** or the pencil beside **Companies**, and the source goes back
    to **Active** by itself.

    **Finished** means the search is done: every company matching those filters has been analyzed, or the
    filters matched none at all. Raising **Total companies** will not restart a finished source, and the
    play button stays disabled. Add a source with wider targeting instead.

    **No Credits**: the workspace ran out of credits partway through this source's search. Add credits or
    wait for your next allotment; a source paused this way is picked up again on its own once the balance
    covers it.
  </Accordion>

  <Accordion title="A source says Active, but its Companies count isn't moving today">
    Hover the chip. If the tooltip says the daily limit was reached, the source has hit its **Daily** cap
    for today and is held until tomorrow rather than paused outright. Raise the daily cap with its own
    pencil if you'd rather not wait.
  </Accordion>

  <Accordion title="Pressing play on a Pending source returns an error">
    The searches behind the source are still being built, and it cannot start until they exist. Wait a
    moment and press play again, or leave it: the page keeps re-checking, and a source usually leaves
    **Pending** without any help.
  </Accordion>
</AccordionGroup>

## Related

<CardGroup cols={2}>
  <Card title="Decision-Maker Targeting: filters explained" icon="users" href="/en/finding-leads/decision-maker-targeting">
    Every targeting filter a source can use (industry, size, locations and roles) and what each one widens or narrows.
  </Card>

  <Card title="Quality gate and match score" icon="sliders-horizontal" href="/en/finding-leads/quality-gate-and-match-score">
    How the score behind a source's optional quality gate actually gets calculated.
  </Card>

  <Card title="Agent statuses and pause options" icon="circle-pause" href="/en/ai-agents/agent-statuses-and-pausing">
    The agent-level pill these per-source statuses roll up into, and the three different scopes a pause can cover.
  </Card>

  <Card title="Tags: how agents, sources and leads connect" icon="tag" href="/en/leads/tags">
    Why adding a source can create a tag on the spot, and what an orphaned source actually is.
  </Card>
</CardGroup>