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

# Agentic capabilities

> The capability map for agents on the Reply.io platform — outcomes, access layers, and where the human boundary sits.

Reply.io is a sales-engagement platform an AI agent can operate end to end: find and manage
contacts, launch multichannel outreach, triage and answer replies, pull performance data, and
program Jason — Reply's AI SDR — as a supervised autopilot. Everything runs behind one scoped
[API key](/api-reference/authentication) across four access layers.

## What an agent can do today

| Outcome                                                          | Status         | Where to start                                                       |
| ---------------------------------------------------------------- | -------------- | -------------------------------------------------------------------- |
| Import and manage contacts, lists, accounts                      | ✅ Available    | [Import and clean contacts](/guides/import-contacts)                 |
| Launch multichannel sequences (email + LinkedIn)                 | ✅ Available    | [Launch multichannel outreach](/guides/launch-outreach)              |
| Manage and optimize running sequences                            | ✅ Available    | [Improve sequences every cycle](/guides/improve-sequences)           |
| Triage the inbox and send approved replies                       | ✅ Available    | [Convert replies into meetings](/guides/convert-replies-to-meetings) |
| Pull reports and per-sequence stats                              | ✅ Available    | [Decide what to scale or fix](/guides/analyze-performance)           |
| Configure Jason AI SDR (knowledge, offers, playbooks, approvals) | ✅ Available    | [Build a supervised AI SDR](/guides/build-an-ai-sdr)                 |
| Event-driven automation via webhooks                             | ✅ Available    | [Automate outbound end-to-end](/guides/automate-outbound)            |
| Validate emails, enforce blacklists, keep lists clean            | ✅ Available    | [Automate outbound end-to-end](/guides/automate-outbound)            |
| Search the 1B+ B2B database for new prospects                    | 🔜 Coming soon | [Find your ideal prospects](/guides/find-ideal-prospects)            |
| Waterfall-enrich contacts (emails, phones, firmographics)        | 🔜 Coming soon | [Find your ideal prospects](/guides/find-ideal-prospects)            |

## The four access layers

All four surfaces operate the same platform — same key, same objects, same limits. They differ in
how an agent consumes them:

```
                    ┌────────────────────────────┐
   MCP client   ──▶ │  MCP · 70 annotated tools  │ ──┐
                    └────────────────────────────┘   │
                    ┌────────────────────────────┐   │      ┌───────────────┐
   Coding agent ──▶ │  Skill · packaged workflows│ ──┤ ───▶ │  Reply.io v3  │
                    │  └─▶ CLI · `reply` command │   │      │   platform    │
                    └────────────────────────────┘   │      └───────────────┘
                    ┌────────────────────────────┐   │
   Any program  ──▶ │  REST API · api.reply.io/v3│ ──┘
                    └────────────────────────────┘
```

* **[MCP](/mcp/overview)** — direct tool access for AI agents in MCP clients.
* **[Skills](/skills/overview)** — outbound expertise and guardrails as markdown packs; the skill
  decides *what* to do, the CLI or MCP does it.
* **[CLI](/cli/overview)** — terminal and automation interface with JSON output and safety flags.
* **[API](/api-reference/introduction)** — low-level programmable integration; the exhaustive surface.

Not sure which fits? [Choose your interface](/agents/choose-your-interface).

## The human boundary today

An agent can operate everything behind the API key. Three things still need a human:

1. **Account creation** — a human signs up (14-day free trial, no credit card required).
2. **API key issuance** — created in the Reply app under **Settings → API Keys** and handed to the
   agent. Keys support granular scopes — see [Authentication](/api-reference/authentication).
3. **Billing** — plan upgrades and credit purchases happen in the Reply app.

## Next steps

<CardGroup cols={2}>
  <Card title="Choose your interface" icon="scale-balanced" href="/agents/choose-your-interface">
    MCP vs Skills vs CLI vs API — decision rules and a comparison table.
  </Card>

  <Card title="Agent safety rules" icon="shield" href="/agents/safety">
    High-stakes gating, scoped keys, and rate-limit etiquette across all surfaces.
  </Card>

  <Card title="Drop-in system prompts" icon="message-code" href="/agents/prompts">
    Ready-to-paste prompts for agents operating Reply.
  </Card>

  <Card title="Guides" icon="map" href="/guides/launch-outreach">
    Task-based guides for every outcome in the table above.
  </Card>
</CardGroup>