Revid

Make a series episode

create_episode

Make a whole episode of a series with Revid's Series Maker, on Autopilot: the server plans the story and shot list, draws each character, place and prop once, generates every scene clip with them so they look the same in every scene, and assembles the episode with music, captions and the series outro. Without seriesId this starts a new series; with a seriesId from list_series it makes that series' next episode with the same cast, places, style, ratio, model and language (do not send those then). Returns a runId: poll get_episode_status every 30 to 60 seconds (a whole episode usually takes 10 to 30 minutes). When it is done, call export_video with its pid, then get_project_status until videoUrl. Credits are charged per reference image and scene clip as they come back, never past maxCredits (default 2500): the run stops with status needs_attention first. Call get_account to check the balance before a long Ultra episode. Requires an active Revid plan.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "prompt": "<prompt>"
    },
    "name": "create_episode"
  }
}

Input Schema

Name	Required	Description
`ratio`	No	Aspect ratio of a new series. Default 9:16.
`style`	No	Visual style of a new series. Default: default (Revid picks from the prompt).
`prompt`	Yes	What the episode is about, or for a new series what the show is: premise, characters, tone, what happens.
`language`	No	Spoken language of a new series (ISO 639-1, e.g. en, fr, es). Default: the prompt's language.
`seriesId`	No	Make this series' next episode (from `list_series`). Omit to start a new series.
`maxCredits`	No	The most this episode may spend. Default 2500.
`videoModel`	No	Video tier of a new series, per 5 s of video: low 15 credits (cheapest), pro 50 (default), ultra 100 (Seedance 2.5, best acting, scenes up to 30 s).
`durationSeconds`	No	Whole episode length in seconds, outro included, in 5 s steps. Omit for Auto.

Output Schema

Name	Required	Description	Default
`ok`	Yes
`data`	No
`error`	No
`method`	Yes
`status`	Yes
`endpoint`	Yes
`retryAfterSeconds`	No

Schema Changelog

Changes observed during successful MCP inspections.

First observedNov 5, 2026

TDQS

A4.8/5.0

Behavior5/5

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

Annotations only flag it as a non-readonly, non-idempotent, open-world write; the description adds the operational facts an agent actually needs: expected duration (10–30 min), credit accrual per reference image and scene clip, the hard `maxCredits` stop that surfaces as `needs_attention` before overrunning, and the requirement of an active Revid plan. It also explains why reusing a `seriesId` matters — cast, places, style, ratio, model and language are inherited for visual consistency.

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?

Front-loaded with the action and the pipeline, then modes, then the post-call polling chain, then cost/safety notes — a logical progression with no filler sentences. It is dense as a single undivided paragraph, so the branching and polling instructions would scan faster as separated clauses, but no sentence is redundant.

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?

For an expensive, long-running, multi-stage generation job this covers everything an agent needs: mode selection, parameter exclusions, return `runId`, polling cadence and expected wait, the export handoff chain, credit exposure and the `needs_attention` failure mode, and the plan prerequisite. An output schema exists, so the absence of a return-value breakdown is not a gap.

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 100%, so the baseline is 3, but the description adds genuinely new semantics: `seriesId` as the switch between new-series and next-episode modes plus the mutual exclusion of the series-shaping params, and what `maxCredits` actually does when reached. It leaves the prompt-shaping nuance (premise vs. episode brief) mostly to the schema field text.

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 specific verb and resource ('Make a whole episode of a series') and immediately details the Autopilot pipeline (plan, character/place/prop reference drawing, scene clips, assembly with music/captions/outro). It also distinguishes its two operating modes by `seriesId`, so an agent can tell it apart from `generate_media`, `edit_project` or `create_character` without opening a schema.

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?

Gives explicit branching rules: omit `seriesId` to start a new series, pass one from `list_series` to continue an existing series, and 'do not send those then' forbids re-sending ratio/style/model/language. It also prescribes the full downstream sequence (`get_episode_status` every 30–60 s, then `export_video` with the `pid`, then `get_project_status` until `videoUrl`) and a pre-flight check via `get_account` for long Ultra runs.

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

All Revid 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?