blockout-mcp
Enables AI agents to control the Blockout previs desktop app for AI filmmaking, allowing staging of 3D worlds, character animation, camera framing, timeline control, and viewport screenshotting through MCP tools.
README
blockout-mcp
MCP (Model Context Protocol) stdio bridge for Blockout — the open-source previs desktop app for AI filmmaking. With this server connected, an AI agent can stage grey-box worlds, drop in choreographed crowds, animate characters, frame and move the camera with classic cinematography presets, scrub the timeline, take viewport screenshots, and drive exports — everything a human does in the app.
Zero dependencies. Node ≥ 18. One file.
Requirements
- The Blockout desktop app must be running with a project open. Get it from the Blockout repo (build from source or use a release DMG).
- On launch, Blockout starts a localhost-only control server on a random port and writes a discovery file to
~/.config/blockout/control.json(random bearer token, mode 0600, removed on quit). This bridge reads that file automatically — there is nothing to configure and no credentials to enter.
Connect
Hermes
Add to ~/.hermes/config.yaml (or install from the Hermes MCP catalog once listed):
mcp_servers:
blockout:
command: "node"
args: ["/absolute/path/to/blockout-mcp/blockout-mcp.mjs"]
Claude Code
claude mcp add blockout -- node /absolute/path/to/blockout-mcp/blockout-mcp.mjs
Any MCP client (generic stdio config)
{ "mcpServers": { "blockout": { "command": "node", "args": ["/absolute/path/to/blockout-mcp/blockout-mcp.mjs"] } } }
Tools (30)
get_state · list_assets · add_entity · move_entity · delete_entity · add_actor_mark · add_camera_mark · clear_camera_marks · set_shot · new_shot · apply_framing · list_camera_moves · apply_camera_move · set_track_subject · list_action_presets · apply_action_preset · list_sequence_styles · spawn_sequence · set_reference · snap_to_ground · set_time · play · stop · screenshot · list_presets · save_preset · apply_preset and more — call get_state first; every tool description carries the scene conventions (meters; +X right, −Z forward; heading 0 faces −Z).
A typical agent session: get_state → add_entity/spawn_sequence → apply_camera_move → screenshot to see the result → set_shot/play.
Security
The control server binds to 127.0.0.1 only, uses a per-launch random bearer token, and the app validates every action against a whitelist. Nothing is exposed off-machine.
License & credit
Apache-2.0 — see LICENSE. Per the NOTICE file, use, forks, and redistribution must credit Sam Wasserman (wassermanproductions.com).
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
Exa MCP Server
模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。