{"slug": "palmier-pro-ai-video-editing-via-mcp", "title": "Palmier Pro: AI Video Editing via MCP", "summary": "Palmier Pro, a macOS video editor built in Swift, now integrates AI agents via MCP (Model Context Protocol), allowing LLMs like Claude or Codex to automate media management, semantic search, timeline manipulation, asset generation, and local analysis using CoreML and SpeechAnalyzer. The tool offers a local MCP server for complex workflows and an in-app chat for quick iterations, transforming the editor into a programmable interface for AI-driven rough cuts and beat-matching.", "body_md": "# Palmier Pro: AI Video Editing via MCP\n\nThe technical stack is lean—built in Swift to avoid the overhead of Electron/Node.js—and leverages native macOS APIs like CoreML and SpeechAnalyzer for local processing.\n\n## Core AI Workflow Capabilities\n\nConnecting [Claude](/en/tags/claude/) or Codex to the local MCP server allows the agent to handle the mechanical \"grunt work\" that usually kills creative momentum. Here is what the agent can actually execute:\n\n**Media Management:** Importing files from URLs or local paths and organizing them into folders.**Semantic Search:** Finding footage via local SigLIP2 embeddings.**Timeline Manipulation:** Performing operations on tracks, clips, and keyframes.**Asset Generation:** Creating music, SFX, captions, and video clips within the editor.**Local Analysis:** Running Silero VAD for silence detection and beat_this for beat-matching.\n\n## Deployment and Setup\n\nSince this is designed for macOS, you get native performance without the webview lag. The integration works in two ways:\n\n1. **Local MCP Server:** Best for complex, automated workflows (e.g., replicating a specific podcast editing style across multiple raw files).\n\n2. **In-App Chat:** Better for quick, low-latency iterations while you have the timeline open.\n\nFor those looking to build a custom AI video workflow, the ability to trigger exports and edits via an LLM agent transforms the editor from a manual tool into a programmable interface. It's particularly effective for rough cuts based on transcriptions or beats, where the AI handles the alignment and the human handles the final creative polish.\n\nIf you are running macOS, it's a great way to test how an LLM agent can manage a heavy professional application without the usual API friction.\n\n[Next Loop Engineering: Stopping Agent Reward-Hacking →](/en/threads/2183/)", "url": "https://wpnews.pro/news/palmier-pro-ai-video-editing-via-mcp", "canonical_source": "https://promptcube3.com/en/threads/2188/", "published_at": "2026-07-23 09:45:32+00:00", "updated_at": "2026-07-23 18:10:08.957942+00:00", "lang": "en", "topics": ["ai-tools", "ai-agents", "artificial-intelligence", "developer-tools"], "entities": ["Palmier Pro", "Claude", "Codex", "CoreML", "SpeechAnalyzer", "SigLIP2", "Silero VAD", "beat_this"], "alternates": {"html": "https://wpnews.pro/news/palmier-pro-ai-video-editing-via-mcp", "markdown": "https://wpnews.pro/news/palmier-pro-ai-video-editing-via-mcp.md", "text": "https://wpnews.pro/news/palmier-pro-ai-video-editing-via-mcp.txt", "jsonld": "https://wpnews.pro/news/palmier-pro-ai-video-editing-via-mcp.jsonld"}}