MCP
Every SLOP TV News story about MCP.
Why AI video generation is turning into an agent loop
A new survey names the runtime around a video model the agent harness, and this week Loova, Runway and ComfyUI shipped the same idea as products.
Vibe says Higgsfield can launch a streaming TV ad in one chat
The integration runs on the Model Context Protocol, so Higgsfield's agent Supercomputer calls Vibe's ad tools directly; financial terms were not disclosed.
How to add Runway Dev to Claude Code, Codex and Cursor
One install wires your coding agent to your Runway Dev account, so it can read your models, prices and credit balance and build the integration itself.
RealityChek adds deepfake checks to ChatGPT and Claude
The company says the DETECT connector runs on the Model Context Protocol, so users can check images, video, audio and documents without leaving a conversation.
Mosaic's Motion agent takes on editing and video understanding
The agent cuts, motion-designs and grades video from a brief, then hands back a project you can finish in Premiere, After Effects or DaVinci Resolve.
This week in media MCPs: Unity ships 32 skills and two servers retire
Runway and ElevenLabs retired their local MCP servers in the same week Unity shipped 32 official skills for coding agents. Every media-creation MCP server and skill pack we track, with the date each figure was read.
Pika puts 40+ creative tools inside ChatGPT's side panel
The plugin, version 0.19.3, syncs finished media back to a Pika library and grants 250 credits after the first generation.
Rushilo ships a $99 video editor that Claude Code and Codex drive
The Mac and Windows app connects to Claude Code, Codex or any MCP client, runs Whisper captions on the machine and charges no AI credits.
Higgsfield Katana builds video edits from a reference in Claude
The tool runs inside Anthropic's Claude through Higgsfield's MCP connector and turns an uploaded clip into editable motion graphics.
invideo's Agentic Extend lengthens a scene by up to 30 seconds
invideo says the extension continues a scene's dialogue, music and sound effects on separate layers the editor can still adjust by hand.
Eloelo opens Dolphin AI, an agentic studio for multi-shot video
The Bengaluru company's platform plans a script into scenes, keeps characters and wardrobes steady between shots, and plugs into Claude and ChatGPT over MCP.
Claude Fable 5.1 is the left brain, GPT-6 Astra the right
Neither model can make a single frame of video. The benchmarks that pull them apart measure language and spatial sense, which is where the old split-brain research drew its line too.
How to build a car ad with Claude Opus 5.5 and Seedance 2.5
JSFILMZ published a walkthrough of just under ten minutes that runs from the Higgsfield connector to two finished ad cuts. Here is every step, and the settings that decide the result.
Higgsfield's Blender add-on generates meshes, rigs and lit scenes
The add-on needs Blender 5.1 or later and runs on Higgsfield's cloud credits, and only some of its seven tools return data you can keep editing.
Invideo says manual editing is dying. The first pass is.
Vendor threads promise the timeline is over, while the tools doing the actual work are aimed at the boring first pass rather than the cut.