JUNKDOGE-JOE/after-effects-mcp

Description and use

**Brief summary**
JUNKDOGE‑JOE/after‑effects‑mcp is an open‑source stack that lets AI agents (Codex, Cursor, Claude, etc.) control Adobe After Effects via a Model Context Protocol (MCP) server and a CEP plugin, exposing 30 native ae.* commands. By standardising the “AI‑to‑tool” interface, it makes it possible to drive real‑world video‑editing workflows with code‑generated instructions.

**Value**
The project bridges the gap between large‑language‑model assistants and production‑grade creative software, turning natural‑language prompts into concrete After Effects actions without manual scripting. This enables rapid prototyping of AI‑driven motion‑graphics pipelines, reduces the need for custom ExtendScript code, and creates a reusable protocol that other tools can adopt.

**Practical adoption path**

1. **Prototype** – Clone the repo, run the MCP server (Node.js) locally, and install the CEP plugin into After Effects. Use the provided CLI or REST endpoint to send simple ae.* commands (e.g., create a composition, add a layer).
2. **Integrate** – Wrap the MCP calls in your preferred AI‑agent framework (e.g., LangChain, AutoGPT). Map high‑level intents (“add a lower‑third with fade‑in”) to the corresponding ae.* calls.
3. **Scale** – Deploy the MCP server as a containerized microservice (Docker/K8s) and expose it to multiple agents or a CI/CD pipeline that generates video assets automatically.
4. **Standardise** – Share the MCP schema with other teams or open‑source projects, allowing different AI models to speak the same “After Effects” language.

**Production readiness**
The repository shows strong recent activity (last commit 2026‑07‑08), a modest but growing community (24 stars, 2 forks), and clear API/CLI entry points, indicating a mature codebase for pilot projects. While the license and security posture still need a final audit, the core components (MCP server, CEP plugin) are stable, documented, and written in JavaScript, making integration straightforward for teams already using Node.js and After Effects. Consequently, it is suitable for a serious production trial, especially for internal automation or proof‑of‑concept deployments.

Use in a workflow

JUNKDOGE-JOE/after-effects-mcp helps connect AI assistants to real tools and data through a standard protocol.

• connect AI agents to tools
• ship Model Context Protocol servers
• standardize integrations

Limitations and review notes

High for an OSS candidate: recent activity, adoption, and ecosystem signals are strong enough for a serious pilot.