> For the complete documentation index, see [llms.txt](https://docs.podigee.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.podigee.com/documentation/start-here/ai-agents.md).

# AI Agents and MCP

Connect AI assistants to Podigee and complete podcast work with clear permission and confirmation controls.

Podigee for AI agents connects compatible AI applications to the podcasts that you manage in Podigee. Your agent can find podcast information, prepare episodes, publish media, import shows, review analytics, check distribution, and manage private-feed access.

You remain in control. The connection uses the Podigee account and permissions that you select. Podigee previews consequential changes before they run and requires confirmation for sensitive actions.

## Choose how to connect

Use the connection that matches your work:

* Use a verified compatible remote MCP client for conversational work. Public first-party ChatGPT and Claude listings will be separate releases.
* Use the `pdg` command-line interface (CLI) when your workflow must read a local audio or video file.
* Connect another compatible Model Context Protocol (MCP) client when you use a different AI application.

MCP is the open standard that lets an AI application use Podigee tools, resources, and guided workflows. You do not need to understand the protocol to use a compatible client.

## What your agent can do

The available work is organized around podcast goals:

1. **Inspect your workspace.** Find podcasts and episodes. Check the connected account, granted access, subscription state, and usage limits.
2. **Prepare and publish episodes.** Create drafts, update metadata, upload local media, monitor encoding, publish now, or schedule publication.
3. **Move existing work.** Import a podcast from an RSS feed and monitor the import until it is complete.
4. **Reuse episode content.** Read transcripts, search what was said, and use generated content assets as source material.
5. **Review performance.** Request podcast or episode analytics within the history available to your plan.
6. **Manage audience access.** Check distribution, submit supported directories, and manage private-feed listeners.

The available actions can depend on your Podigee plan, account access, selected permissions, and the state of the podcast or episode.

## How a safe change works

A typical change has five stages:

1. Your agent reads the current Podigee state.
2. Your agent prepares the requested change.
3. Podigee returns a preview for you to review.
4. You confirm the exact change.
5. Your agent reads the result and verifies the outcome.

Long-running work, such as publication or import, returns an operation ID. Your agent can use that ID to monitor, resume, or cancel eligible work without starting a duplicate operation.

## Local media uses the pdg CLI

AI applications do not send large audio or video files inside MCP messages. Use the `pdg` CLI or a compatible upload client to transfer local media directly to Podigee. The agent can then prepare the episode, start encoding, and monitor publication.

An agent can manage an episode that already has media without uploading the file again.

## Start with a read-only check

Begin with [Connect and run your first read-only check](/documentation/start-here/ai-agents/start-here/connect-and-run-your-first-read-only-check.md). This confirms the connected account and lists the podcasts that the connection can access before you grant write permission or publish media.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.podigee.com/documentation/start-here/ai-agents.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.