Digital Maker AI

create_podcast

Start an Animated Podcast episode: two AI hosts with two different voices talking through a topic over a held studio picture, with captions. Returns a job id at once; use check_video for progress and the link.

SPENDS REAL MONEY: one minute from the user's monthly pool per minute of episode. Premium up to 30 minutes, Ultimate up to 60. Agree topic, length, tone, the two hosts (names and voices, which must differ) and the studio with the user first, and call once per request on their own direct request only. The studio is an image_url from preview_podcast_studio the user approved, the name of a saved studio from list_podcast_studios (which also brings that studio's host names, looks and voices unless you give others), or a public link to the user's own finished studio picture. Voices are the built-in ones from list_video_options (not cloned voices). No background music on podcasts. SHOWS: if show_name is one of the user's saved shows, the settings of its last episode fill in anything you do not pass, so a repeat episode only needs a topic; save_show remembers this episode's set-up under show_name. After it is finished: make_podcast_thumbnail for a thumbnail with these hosts, podcast_youtube_details for the YouTube title, description, chapters and tags, and edit_podcast to change lines, voices, captions, sound bar or studio.

Input Schema

TableJSON Schema

Name	Required	Description
`tone`	No	friendly = warm, curious chat (default); educational = one explains, one asks; debate = two sides, kept respectful; funny = quick banter with substance; storytelling = one tells, one reacts.
`topic`	Yes	What the episode is about.
`host_a`	No	Left host.
`host_b`	No	Right host.
`studio`	No	Or: the name of a saved studio (list_podcast_studios).
`minutes`	No	Length in minutes. Premium 1 to 30, Ultimate 1 to 60. Defaults to 5.
`like_cta`	No
`soundbar`	No	The bars that bounce with the voices. classic = top centre (default); top = full-width along the top edge; hosts = a set above each host, only the talker's light up; edge = full-width along the bottom; corner = small, top-left; off = none.
`save_show`	No	Remember this episode's hosts, voices, studio and settings under show_name, so the next episode only needs a topic.
`show_name`	No	Optional show name the hosts use.
`comment_cta`	No
`subscribe_cta`	No
`subtitle_style`	No	Captions. Defaults to clean; none turns them off.
`studio_image_url`	No	image_url from preview_podcast_studio after the user approved it, or a public link (https) to the user's own finished studio picture with the two hosts in it.

Schema Changelog

Changes observed during successful MCP inspections.

First observedOct 17, 2026

TDQS

A4.8/5.0

Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations, the description carries the full burden and does so: it warns 'SPENDS REAL MONEY' with the exact cost model (minutes from the monthly pool, Premium 30 / Ultimate 60 caps), discloses the async behavior (returns a job id at once), states hosts' voices must differ, that only built-in voices (not clones) are allowed, and that podcasts have no background music.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Long, but the highest-stakes information (the money warning and the once-per-request rule) is front-loaded and the rest is dense prerequisite/lifecycle guidance rather than filler. For a 14-param tool with nested hosts and multiple dependencies, the length is largely earned; a slight trimming pass on follow-up tooling would help.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the complexity (14 parameters, nested host objects, no output schema, no annotations), the description covers cost, async return, prerequisites, constraint on voices, show inheritance, and the post-generation workflow. An agent has everything needed to call it correctly without opening other tools.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 79% (baseline 3), and the description adds real meaning on top: it explains the three legitimate sources for studio (approved preview image_url, saved-studio name, or public link), the origin of voices (list_video_options built-ins), and the show_name/save_show inheritance behavior. Only minor gaps (e.g., like_cta/comment_cta/subscribe_cta semantics) remain undocumented.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

Opens with a concrete verb+resource ('Start an Animated Podcast episode') and specifies the distinguishing shape: two AI hosts, two different voices, a topic, a held studio picture, captions. It is clearly separable from create_video and animate_scenes, and it names its follow-up tools (check_video, edit_podcast) for lifecycle clarity.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly states when and how to call: agree topic/length/tone/hosts/studio with the user first, 'call once per request on their own direct request only.' It routes to the correct prerequisite tools (preview_podcast_studio, list_podcast_studios, list_video_options) and to check_video for progress, leaving little to inference.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Digital Maker AI tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?