← Back to Skills Marketplace
ergopooka

Assimilate MCP

by Alexander MacLean · GitHub ↗ · v1.0.0
cross-platform ⚠ suspicious
617
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install assimilate-mcp
Description
Control Assimilate Live FX / SCRATCH — professional color grading, compositing, and virtual production software — via MCP. 88 tools across 14 categories.
README (SKILL.md)

Assimilate MCP

Control Assimilate Live FX / SCRATCH — professional color grading, compositing, and virtual production software — via MCP. Complete 1:1 integration of the Assimilate REST API with 88 tools across 14 categories.

Prerequisites

Setup

MCPorter

mcporter config add assimilate --command npx --args '["-y", "assimilate-mcp"]'
mcporter list assimilate

Claude Desktop

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "assimilate": {
      "command": "npx",
      "args": ["-y", "assimilate-mcp"]
    }
  }
}

Claude Code

claude mcp add assimilate -- npx -y assimilate-mcp

Configuration

Flag Env Var Default Description
--host ASSIMILATE_HOST 127.0.0.1 Live FX host
--port ASSIMILATE_PORT 8080 REST API port
--key ASSIMILATE_KEY — Authorization key
--timeout ASSIMILATE_TIMEOUT 30000 HTTP timeout (ms)

Example with custom port:

{
  "mcpServers": {
    "assimilate": {
      "command": "npx",
      "args": ["-y", "assimilate-mcp", "--port=9090"]
    }
  }
}

Tools (88)

Category Count Key Tools
System 8 get_system check_connection list_users select_user
Projects 7 list_projects enter_project create_project
Groups 9 list_groups get_current_group create_group
Constructs 10 list_constructs create_construct enter_construct
Slots 5 list_slots get_slot set_slot create_slot
Versions 5 list_versions get_version set_version
Shots 7 get_shot set_shot create_shot import_media
Inputs 4 get_inputs get_input set_input
Color Grading 5 get_grade set_grade get_framing set_framing
Player 8 enter_timeline set_playmode enter_shot exit_player
Render 10 start_render stop_render get_render_status
Outputs 6 list_outputs create_output set_output
Snapshots 2 render_snapshot get_shot_metadata
Files 2 list_directory find_media

Usage Examples

Talk to your AI assistant in natural language:

  • "What projects are available?"
  • "Import the ARRIRAW files from /Volumes/Shuttle/Day_14"
  • "Warm up the gamma on this shot"
  • "Set up ProRes 4444 output and render the timeline"
  • "Take a snapshot of this frame"

MCPorter CLI

mcporter call assimilate.check_connection
mcporter call assimilate.list_projects
mcporter call 'assimilate.enter_project(name: "Commercial_Nike_Q3")'
mcporter call assimilate.get_grade

Remote Access

Live FX accepts connections on localhost by default. For remote machines, use an SSH tunnel:

ssh -f -N -L 8080:127.0.0.1:8080 user@livefx-host

Links

Usage Guidance
This skill appears to be what it claims (a npx-based MCP adapter for Assimilate), but proceed with caution: 1) The SKILL.md references an ASSIMILATE_KEY (authorization token) but the skill metadata does not declare required env vars — treat this as an incoherence and avoid supplying sensitive keys until you verify the package. 2) npx will pull and execute code from npm at runtime; inspect the package on npm and the linked GitHub repository (check publisher, recent commits, and package contents) before running. 3) Because the tool can access local filesystem paths and network endpoints (it imports media and speaks to a REST API), run it first in an isolated environment or sandbox and audit network calls. 4) Prefer pinning a specific vetted package version, or install the package locally and review its source, rather than allowing ad-hoc npx downloads. If you need help auditing the npm package or confirming the GitHub repo matches the published package, gather the package tarball or repository link and re-run an evaluation.
Capability Analysis
Type: OpenClaw Skill Name: assimilate-mcp Version: 1.0.0 The skill bundle is classified as suspicious due to its reliance on an external npm package (`assimilate-mcp`) executed via `npx`, which introduces a supply chain risk. The underlying tool, as described in `SKILL.md`, grants the AI agent broad filesystem access (e.g., `list_directory`, `find_media`, `import_media`) and network access to a local REST API, potentially handling an authorization key. While these capabilities are necessary for the stated purpose of controlling media software, they represent significant high-risk behaviors that could be exploited through prompt injection or if the external package itself were compromised, without clear evidence of intentional malice within the skill bundle itself.
Capability Assessment
✓ Purpose & Capability
Name, description, and declared binaries (npx) match the actual behavior: the skill runs an npm package (assimilate-mcp) that proxies commands to the Assimilate REST API. Required tooling is appropriate for the stated purpose.
ℹ Instruction Scope
SKILL.md stays within the domain of controlling Assimilate (listing tools, configuring host/port, examples). It does reference operations that access local paths (e.g., importing media from /Volumes/...), and advises SSH tunneling for remote access — these are expected for a media control tool but mean the invoked package will be able to read local files and network endpoints. Also, the docs reference an ASSIMILATE_KEY and other env vars that the runtime will use.
ℹ Install Mechanism
Installation is via the public npm package 'assimilate-mcp' (executed with npx). This is a normal distribution method for this type of tool, but it implies code will be fetched and executed from the npm registry at runtime; the SKILL bundle contains no code to audit locally. Verify the npm package and GitHub repo prior to allowing execution.
⚠ Credentials
Registry metadata lists no required env vars, but SKILL.md documents ASSIMILATE_HOST, ASSIMILATE_PORT, ASSIMILATE_KEY, and ASSIMILATE_TIMEOUT — notably an authorization key (ASSIMILATE_KEY). The declared metadata should have listed required credentials. Requesting a local REST API key is plausible for this integration, but the mismatch (no required env vars declared) is an incoherence and increases risk because sensitive credentials could be supplied without the skill explicitly declaring them.
✓ Persistence & Privilege
always is false and the skill does not request elevated persistence or modify other skills. It suggests adding configuration to client config files but does not claim to alter system-wide settings automatically. Normal agent autonomy remains enabled (default).
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install assimilate-mcp
  3. After installation, invoke the skill by name or use /assimilate-mcp
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
Initial release of assimilate-mcp. - Control Assimilate Live FX / SCRATCH via MCP with 88 tools across 14 categories. - Provides full 1:1 integration with the Assimilate REST API. - Supports project, color grading, render, player, snapshot, and file management commands. - Includes setup guides for MCPorter, Claude Desktop, and Claude Code. - Highly configurable with command-line flags and environment variables.
Metadata
Slug assimilate-mcp
Version 1.0.0
License —
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Assimilate MCP?

Control Assimilate Live FX / SCRATCH — professional color grading, compositing, and virtual production software — via MCP. 88 tools across 14 categories. It is an AI Agent Skill for Claude Code / OpenClaw, with 617 downloads so far.

How do I install Assimilate MCP?

Run "/install assimilate-mcp" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Assimilate MCP free?

Yes, Assimilate MCP is completely free (open-source). You can download, install and use it at no cost.

Which platforms does Assimilate MCP support?

Assimilate MCP is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Assimilate MCP?

It is built and maintained by Alexander MacLean (@ergopooka); the current version is v1.0.0.

More Skills
self-improving agent
Captures learnings, errors, and corrections to enable continuous improvement. Use when: (1) A command or operation fails unexpectedly, (2) User corrects Clau...
⬇ 463815 · by pskoett
Skill Vetter
Security-first skill vetting for AI agents. Use before installing any skill from ClawdHub, GitHub, or other sources. Checks for red flags, permission scope, and suspicious patterns.
⬇ 259802 · by spclaudehome
Polymarket
Query Polymarket prediction markets. Check odds, find trending markets, search events, track price movements.
⬇ 232503 · by joelchance
Self-Improving + Proactive Agent
Self-reflection + Self-criticism + Self-learning + Self-organizing memory. Agent evaluates its own work, catches mistakes, and improves permanently. Use when...
⬇ 200680 · by ivangdavila
Github
Interact with GitHub using the `gh` CLI. Use `gh issue`, `gh pr`, `gh run`, and `gh api` for issues, PRs, CI runs, and advanced queries.
⬇ 191345 · by steipete
ontology
Typed knowledge graph for structured agent memory and composable skills. Use when creating/querying entities (Person, Project, Task, Event, Document), linkin...
⬇ 190333 · by oswalpalash

💬 Comments