21,636 mods in this category, of every kind an
agent can take. Each one carries what it costs per session, what the
scan found, and whether it is the original.
Transcribe public videos and audio (YouTube, TikTok, Instagram, Facebook, Pinterest, Google Drive) into accurate, timestamped text — then summarize, translate or export SRT/VTT subtitles without leaving Claude Code.
★not rated 0 2mo agoA
tokens not measured
originalMIT
Suggest what the owner should watch, read, listen to or cook next — either from the pile they already queued and started, or from outside it. Use when they ask what to watch tonight, what to read next, what to pull off their own shelf, whether they would like a specific thing, or want their stalled shows and abandoned…
Generate or edit images with GPT-Image 2.5 (default $0.03/image flat VIP routes) and Nano Banana 2 via direct HTTP API calls. Replaces MCP image-generate-mcp-remote.
Generate images and video on a 2DAI account through the 2dai-mcp-server tools — pick the right preset, budget credit, keep a character consistent across shots, assemble a short film, and avoid the mistakes agents make on their first session.
Find the right Kavel AI image or video tool for a job, get a model-tuned prompt, and open it ready to generate — hairstyle changer, figurine, pet portrait, Nano Banana 2, Veo 3.1 and more.
★not rated 0
changed 26d agoA
tokens not measured
originalMIT
Transcribe audio, video, and podcasts with Pepys directly from Claude Code – diarization, timed SRT/VTT captions, paste-a-link ingestion, and transcript search.
★not rated 0 3d agoA
tokens not measured
originalMIT
A browser task for downloading specified, existing MiniMax music works as watermark-free MP3 files from a logged-in creator page and checking the local files. MiniMax is a platform for creating music.
Give Your Agent A Voice: x402 pay-per-call speech with 20 voices, 10 personas, 31 languages. Runs locally from the @forgemeshlabs/voice-mcp npm package.
★not rated 0 7d agoA
tokens not measured
originalMIT
Generate and edit images, remove backgrounds, upscale, and create sprite sheets through DreamLayer. Runs locally from the @dreamlayer/mcp npm package. Needs 1 environment variable to run.
★not rated 0 22d agoA
tokens not measured
originalMIT
MCP server that gives an AI assistant eyes for video, via the vidgrid render API. Runs locally from the vidgrid-mcp Python package. Needs 1 environment variable to run.
★not rated 0 1mo agoASocket: warn
tokens not measured
originalMIT
Use the gpt-tools MCP server's gptsearch, gptsearchbatch, gptimagegen, and gptimagegenbatch tools with appropriate prompt lengths, batching, and output paths. Load before calling those tools. Does not apply to Codex's native web search or image generation tools.
MCP server "unsplash" as configured in litoondev/claude-talk-to-figma-mcp-main. Runs locally from the @unsplash/mcp-server npm package. Needs 2 environment variables to run.
★not rated 0 1mo agoA
tokens not measured
originalMIT
Default local image-to-3D workflow using Hunyuan3D-2mv, three AutoRemesher output profiles, Paint, and GPU texture baking. Create serial asset batches, gate four-view cutouts, review and repair every profile, and deliver textured models through codex3dmodelshunyuanmv.
Edit real footage from a conversation: cut silences and filler words, add word-timed captions, reframe to 9:16, add b-roll and music, turn long videos into shorts, record website demos, preview, and export the final MP4.
★not rated 0
changed 2d agoA
tokens not measured
originalMIT
MCP server "media-downloader" as configured in AlexanderChen5966/DownloadVideoPythonProject. Launched with /bin/bash /Users/alexander/PycharmProjects/DownloadVideoPythonProject/start_mcp..
Tarot readings from a real shuffled deck: the Familiar Tarot MCP server plus a skill that draws through it and never invents cards.
★not rated 0 1mo agoA
tokens not measured
originalMIT
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: