1. Conduid
  2. Gaming
  3. Kitsu MCP Server
MCP server · Gaming

Kitsu MCP Server

MCP server for Kitsu production management — connect AI assistants to your animation/VFX/game pipeline

Unclaimed gaming
34Low

Scored 6 months ago · breakdown

About Kitsu MCP Server

Kitsu MCP Server is an MCP server in the Gaming category: mCP server for Kitsu production management — connect AI assistants to your animation/VFX/game pipeline.

Install

From source
git clone https://github.com/INGIPSA/kitsu-mcp-server

This server has no ConduID identity, so agent calls to it are not receipted. Pin the version you install and review the source before granting it credentials.

Ask AI

✦

Ask AI about Kitsu MCP Server

Powered by Claude · Grounded in docs

I know everything about Kitsu MCP Server. Ask me about installation, configuration, usage, or troubleshooting.

Security checks

  • ✓README presentThe repository ships documentation.

5 checks have not been run yet.

README

Kitsu MCP Server

An MCP (Model Context Protocol) server that connects AI assistants like Claude to Kitsu — the open-source production management tool for animation, VFX, and game studios.

Built with FastMCP and the Gazu Python client.

Features

95 tools covering the full Kitsu production pipeline:

Category Tools
Projects List, overview, stats, create, close, add/remove task types
Assets List, details, create, update, delete, update custom data, asset types, CSV import/export
Shots & Sequences List, details, create, update, delete, update custom data, batch create, CSV import/export, OTIO import
Tasks List, details, create, batch create, assign, unassign, start task, submit for review, update custom data, update status, set estimates, delete
Time Tracking Add, set, get time spent, get time spent range
Comments Add comments, reply to comment, acknowledge comment, list comment history
Previews Upload, publish, set main, list per task, get URLs, download, update annotations
Attachments List attachment files per task
Casting Get/set shot casting, get asset casting
Team & People List team, update roles, add/remove members, person tasks, create person, invite, departments
Playlists List, create, add entities, remove entities, build movie
Budgets List, create budgets and budget entries
Concepts List, create concepts
Edits List, create, details, update, delete, list previews
Scenes List, create scenes
Metadata List and add custom metadata descriptors
Notifications List, mark all as read, subscribe/unsubscribe from tasks
Other Search, task statuses, daily progress report

Edit / Montage Review Workflow

Edits are full-length montages or sequences (like animatics, trailers, or final cuts) — the Kitsu equivalent of Frame.io's file-based review. Upload a complete video with audio and review it with annotations, comments, and version tracking.

1. create_edit("My Project", "Trailer_v03")
2. create_task("My Project", "Trailer_v03", "Edit", entity_type="edit")
3. upload_preview(task_id, "/path/to/trailer_v03.mp4", comment="v3 with music")

The team can then review the full montage in Kitsu's player with drawing/annotation tools and timecoded comments.

Requirements

  • Python 3.10+
  • A running Kitsu instance (self-hosted or cloud)
  • A Kitsu user account with appropriate permissions

Installation

git clone https://github.com/INGIPSA/kitsu-mcp-server.git
cd kitsu-mcp-server
pip install -r requirements.txt

Configuration

The server requires three environment variables:

Variable Description Example
KITSU_HOST Your Kitsu API URL https://kitsu.yourstudio.com/api
KITSU_USER Login email artist@yourstudio.com
KITSU_PASSWORD Login password your-password

Usage

With Claude Code

Add to your Claude Code MCP config (~/.claude/mcp.json or project-level .mcp.json):

{
  "mcpServers": {
    "kitsu": {
      "command": "python",
      "args": ["path/to/kitsu-mcp-server/server.py"],
      "env": {
        "KITSU_HOST": "https://kitsu.yourstudio.com/api",
        "KITSU_USER": "your-email@example.com",
        "KITSU_PASSWORD": "your-password"
      }
    }
  }
}

With Claude Desktop

Add the same config block to Claude Desktop's settings under MCP servers.

Standalone

export KITSU_HOST="https://kitsu.yourstudio.com/api"
export KITSU_USER="your-email@example.com"
export KITSU_PASSWORD="your-password"

python server.py

Example prompts

Once connected, you can ask your AI assistant things like:

  • "List all open projects in Kitsu"
  • "Create shots SH010 through SH200 in sequence SQ01"
  • "Upload this render as a preview for the lighting task on SH020"
  • "What happened in the project in the last 24 hours?"
  • "Assign the animation task on SH030 to anna@studio.com"
  • "Remove anna@studio.com from the modeling task on Capsule"
  • "Add 4 hours of work on this task for today"
  • "Export all shots to CSV"
  • "Create a new playlist and add all shots from SQ01"
  • "Create an edit called Trailer_v01 and upload the montage for review"
  • "Show me the details and previews for edit Animatic_v2"

Available tools

Read

  • list_projects — List all open projects
  • get_project_overview — Project overview with team, task types, asset types
  • get_project_stats — Task statistics grouped by status
  • list_assets — List assets (optionally filtered by type)
  • get_asset_details — Asset details with tasks
  • list_sequences — List sequences
  • list_shots — List shots (optionally filtered by sequence)
  • get_shot_details — Shot details with tasks and casting
  • list_my_tasks — Tasks assigned to logged-in user
  • list_tasks_for_entity — Tasks for a specific asset, shot, or edit
  • get_task_details — Task details with comment history
  • get_time_spent — Time entries for a task
  • list_comments — Comments on a task
  • get_shot_casting — Asset breakdown for a shot
  • get_asset_casting — Shots an asset appears in
  • list_team_members — Team members in a project
  • get_person_tasks — Tasks for a specific person
  • list_departments — All departments
  • list_playlists — Playlists in a project
  • list_task_statuses — Available task statuses
  • list_notifications — Recent notifications
  • list_preview_files_for_task — All preview files for a task
  • list_attachment_files_for_task — All attachment files for a task
  • get_preview_file_url — Direct image/video URLs for a preview
  • get_time_spents_range — Time spent by person over a date range
  • search — Search entities by name
  • daily_progress_report — Activity summary for last N hours
  • list_concepts — Concepts in a project
  • list_edits — Edits in a project
  • get_edit_details — Edit details with tasks and previews
  • list_previews_for_edit — All preview files for an edit
  • list_scenes — Scenes in a project
  • list_metadata_descriptors — Custom metadata fields
  • get_budgets — Budgets for a project
  • get_budget_entries — Entries in a budget
  • export_assets_csv — Export assets to CSV
  • export_shots_csv — Export shots to CSV

Create

  • create_project — Create a new project
  • create_asset_type — Create an asset type
  • create_asset — Create an asset
  • create_episode — Create an episode
  • create_sequence — Create a sequence
  • create_shot — Create a shot
  • create_task — Create a task for an asset, shot, or edit
  • create_person — Create a new user
  • invite_person — Send invitation email to a user
  • create_department — Create a department
  • create_playlist — Create a playlist
  • create_budget — Create a budget
  • create_budget_entry — Create a budget entry
  • create_concept — Create a concept
  • create_edit — Create an edit (montage/sequence for review)
  • create_scene — Create a scene
  • add_metadata_descriptor — Add a custom metadata field

Update

  • update_task_status — Change task status with optional comment
  • assign_task — Assign a person to a task
  • unassign_task — Remove a person from a task (or clear all assignees)
  • set_task_estimate — Set time estimate for a task
  • add_comment — Add a comment to a task
  • set_shot_casting — Set which assets appear in a shot
  • update_asset — Update asset description/metadata
  • update_shot — Update shot description/frames/metadata
  • update_edit — Update edit name/description
  • update_team_member_role — Change a team member's role
  • add_time_spent — Add time spent on a task
  • set_time_spent — Set time spent on a task
  • add_person_to_department — Add person to department
  • remove_person_from_department — Remove person from department
  • add_entity_to_playlist — Add shot/asset to playlist
  • add_team_member — Add a person to a project team
  • set_main_preview — Set a preview as the main thumbnail
  • start_task — Start a task (set to WIP)
  • submit_for_review — Submit task for supervisor review (WFA)
  • reply_to_comment — Threaded reply to a comment
  • acknowledge_comment — Mark a comment as acknowledged
  • update_asset_data — Update custom metadata on an asset
  • update_shot_data — Update custom metadata on a shot
  • update_task_data — Update custom metadata on a task
  • update_preview_annotations — Add/edit/remove annotations on a preview
  • mark_all_notifications_as_read — Clear all notifications
  • subscribe_to_task — Subscribe to task notifications
  • unsubscribe_from_task — Unsubscribe from task notifications
  • remove_entity_from_playlist — Remove entity from a playlist
  • add_task_type_to_project — Add a task type to a project
  • remove_task_type_from_project — Remove a task type from a project

Batch

  • batch_create_shots — Create multiple shots at once (e.g. SH010-SH200)
  • batch_create_tasks — Add a task type to all shots/assets in a group

Preview

  • upload_preview — Upload a preview file to a task
  • publish_preview — Status + comment + preview in one step
  • download_preview_file — Download a preview file to local path

Import/Export

  • import_assets_csv — Import assets from CSV
  • import_shots_csv — Import shots from CSV
  • import_otio — Import timeline from OpenTimelineIO

Dangerous (require confirm=True)

  • delete_shot — Delete a shot
  • delete_asset — Delete an asset
  • delete_task — Delete a task
  • delete_edit — Delete an edit
  • close_project — Close/archive a project
  • build_playlist_movie — Build a movie from playlist

Changelog

v0.4.0 (2026-04-13)

  • Task workflow: start_task (WIP), submit_for_review (WFA), reply_to_comment, acknowledge_comment
  • Custom data: update_asset_data, update_shot_data, update_task_data — set custom metadata dicts
  • Preview files: get_preview_file_url, download_preview_file, update_preview_annotations, list_preview_files_for_task
  • Attachments: list_attachment_files_for_task
  • Schedule: get_time_spents_range — query time entries over date ranges
  • Notifications: mark_all_notifications_as_read, subscribe_to_task, unsubscribe_from_task
  • Playlist: remove_entity_from_playlist
  • Project: add_task_type_to_project, remove_task_type_from_project

v0.3.0 (2026-04-12)

  • Edit entity support: Full CRUD for Edits — create tasks on edits, get details with previews, update, delete
  • Unassign task: Remove specific people or clear all assignees from a task
  • Edit review workflow: Upload full montages/animatics for team review with annotations (Frame.io-like)

v0.2.0

  • Team management: add/remove members, update roles, invite users
  • Departments: create, add/remove people
  • Metadata descriptors
  • Budget management
  • Concepts, scenes, edits (basic create/list)

v0.1.0

  • Initial release with 60+ tools covering projects, assets, shots, tasks, previews, playlists

License

MIT License — see LICENSE.

Credits

  • Kitsu by CGWire
  • Gazu Python client by CGWire
  • MCP by Anthropic

README copy captured when this server was indexed; it may lag the repository. The original is authoritative.

Questions

About Kitsu MCP Server

How do I install Kitsu MCP Server?

No package has been published for this server, so install it from source: clone the repository and follow its README.

Is Kitsu MCP Server safe to use with an AI agent?

Its trust score is 34 out of 100 (low). It passes 1 of the 1 static security checks run so far. It has no ConduID identity yet, so agent calls to it are not receipted.

Is Kitsu MCP Server still maintained?

Conduid hasn't recorded a commit date for this repository yet. Check the repository directly for recent activity.