Skip to content
Squall Studio logo

Squall Studio MCP server

Which MCP servers are Squall Studio's own, from the official registry and from asking its own domain, whether they answer, and what they offer. A program checked each of these, nobody copied them from a product page.

read 2026-10-25

← Back to Squall Studio

Whose server it is
  • Registry namecom.squallstudio/squall-studioverified in the MCP registry

The registry name sits under squallstudio.com. The registry only lets a publisher use that namespace after proving control of the domain through DNS or HTTP.

When we asked itchecked 2026-10-25
  • Stateanswerschecked 2026-10-25
  • Tools30
  • Server calls itselfstory-weaver-pro
  • Server version1.0.0
  • Protocol version2025-06-18

On 2026-10-25 the server answered initialize and tools/list without a sign-in. First answer in 224 ms.

Should an agent connect itNot proven either way yet
  • Its tools do every use case we checked: turn a story idea or script into episodes; generate consistent characters and locations for a video series.
  • Checked on 2 of the 7 days it takes to judge reliability.

Once connected

  • For "turn a story idea or script into episodes", the likely first call is generate_script.
  • update_shot, update_location, upload_shot_continuity_reference, update_shot_continuity_reference, remove_shot_continuity_reference change data without a sign-in: anyone with the address can call them.

Said by rules from the measurements above. The likely first call is a model's pick from the tool list; read its description before relying on it.

Measured dailydaily checks: 2, the first on 2026-10-25
  • INITE MCP scoreafter a weeka score needs 7 daily checks: 2 so far
  • What an agent can do with it2 of 2 use cases
  • Answeredmeasuring2 of the 7 daily checks it takes to print a share
  • Median first answer267 ms
  • Space in an agent's context≈ 6,652 tokensthe definitions of all its tools (30), which an agent carries on every turn
  • Tool definitionsA · 99 of 100
  • Typed results0 of 30 toolsdeclared outputSchema: results come typed, read once
  • Parameters without a description: in 2 of 30

What an agent can do with it

  • can do · turn a story idea or script into episodes
  • not sure · plan shots with framing, camera and timing
  • can do · generate consistent characters and locations for a video series

Each of the product's use cases, put to Jev against this server's tool list: can an agent do it with these tools?

Worth a look

  • Tools that change data, on a server that answers without a sign-in: update_shot, update_location, upload_shot_continuity_reference, update_shot_continuity_reference, remove_shot_continuity_reference

A program checks the server once a day: initialize and tools/list, no model. Tokens are the definitions' length at about four characters a token. The grade averages six checks of every tool's name, description, schema and annotations; it says how well the tools are described, not whether they work. The score weighs what was measured and nothing else: answering (25%), doing the product's job (25%), definitions (20%), context (10%), use (10%) and how readily a client connects (10%). A part that was not measured is left out, and a hidden instruction caps the score at 40. The state of vendors' MCP servers: method and numbers →

Tools it offers30 tools · read 2026-10-25
  • list_projects

    List all projects owned by the authenticated user.

  • get_project

    Get one project's summary and current workflow step.

  • get_project_episodes

    List the episodes in a project.

  • get_episode

    Get one episode record.

  • get_episode_editor

    Get the full episode tree in one call: project, episode, outline, script, segments, shots, characters, looks, locations and jobs. Prefer this over several narrower calls.

  • get_episode_shots

    List the shots in an episode.

  • list_episode_segments

    List the video segments in an episode.

  • get_scene_assets

    Get one scene with its shots, characters, locations and continuity references.

  • get_project_script

    Get a project's full script package.

  • get_project_characters

    List a project's characters.

  • get_character

    Get one character record.

  • get_character_looks

    List a character's looks.

  • get_project_locations

    List a project's locations.

  • get_location

    Get one location record.

  • get_asset_library

    List generated assets available to a project.

  • get_job

    Get one generation job's status and output. Use this to poll a job returned by a generate_* tool.

  • list_project_jobs

    List generation jobs for a project, newest first.

  • list_shot_continuity_references

    List the continuity reference images attached to one shot. get_scene_assets also returns these, but only for a whole scene — use this when you need the asset ids, for example to remove one or update its guidance. Each reference's saved…

  • generate_script

    Write a full multi-episode script from a story premise. This CREATES A NEW PROJECT — it does not edit an existing one, and it takes no project_id. While the job runs, job.output carries live progress (phase, completedEpisodes…

  • generate_shot_clip

    Generate a video clip for a single shot (episode_shots row). Spends credits. Returns a job id — poll it with get_job.

  • generate_character_image

    Generate a reference image for a character. Spends credits. Returns a job id — poll it with get_job.

  • generate_location_image

    Generate a reference image for a location. Spends credits. Returns a job id — poll it with get_job.

  • generate_shot_background_image

    Generate an empty background plate for one short shot. Both reference inputs steer what the plate itself looks like — layout, prop identity, materials and lighting — and are unrelated to upload_shot_continuity_reference, which guides clip…

  • generate_episode_music

    Generate a background music track for an episode. Requires the episode to already have an assembled video. Spends credits. Returns a job id — poll it with get_job.

  • update_shot

    Overwrite fields on one shot. Read the shot first (get_episode_shots or get_episode_editor), decide the change yourself, then send ONLY the fields you changed — omitted fields are left untouched. To change dialogue, edit…

  • update_location

    Overwrite fields on one location. Read it first (get_project_locations or get_location), then send ONLY the fields you changed — omitted fields are left untouched. IMPORTANT: a location is usually shared by several scenes across episodes…

  • upload_shot_continuity_reference

    Attach an image you produced locally to a shot as a continuity reference. A continuity reference is fed into this shot's next clip generation as a layout, prop, wardrobe and screen-direction guide — it is NOT used as the shot's opening…

  • attach_shot_continuity_reference

    Attach an image that already exists in this project to a shot as a continuity reference. A continuity reference is fed into this shot's next clip generation as a layout, prop, wardrobe and screen-direction guide — it is NOT used as the…

  • update_shot_continuity_reference

    Set or clear the guidance on one continuity reference already attached to a shot — the note telling the video model how to use that image. List references first with list_shot_continuity_references to get the asset id. Like attaching…

  • remove_shot_continuity_reference

    Detach one continuity reference from a shot. List them first with list_shot_continuity_references to get the asset id. This deletes the reference asset permanently and cannot be undone.

How to connect

Add the server to your MCP client config. Claude Desktop, Cursor and most other clients read this shape:

{
  "mcpServers": {
    "squall-studio": {
      "type": "http",
      "url": "https://squallstudio.com/api/mcp"
    }
  }
}

Add it

Claude Code, in a terminal

claude mcp add --transport http squallstudio https://squallstudio.com/api/mcp

Add to CursorAdd to VS Code

claude.ai, ChatGPT and other clients: add https://squallstudio.com/api/mcp as a custom connector in the settings.

Each line is something a program found when it went looking: the registry entry, the server's own answer, a spec that parses. A missing line means we did not find it, not that it does not exist.