Data and AI

14,252 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.

vision

1009

visianlee/vision-mcp

Skill Claude CodeCodex

A vision bridge that sends images to supported visual-language services and returns text descriptions to a text-only model. It includes tools for image analysis, text extraction, image comparison, error screenshots, diagrams, and interface reviews.

not rated 4 4mo ago A 80 tokens original MIT

karpathy-guidelines

1010

Shareed2k/honey

Cursor rule Cursor

Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.

not rated 4 19d ago A 571 tokens copy · 100% MIT

docs-mcp

1011

JaxsonWang/docs-mcp

MCP server Claude CodeCodexCursor +2

Local-only Node.js CLI + MCP server for ingesting Markdown docs and exposing retrieval tools. Runs locally from the @jaxsonwang/docs-mcp npm package.

not rated 4 10mo ago A tokens not measured original MIT

crocodile

1012

nazmiefearmutcu/Crocodile

MCP server Claude CodeCodexCursor +2

Deep Learning Framework & Workload Management For On-premise Personal Machines. Runs locally from the crocodile Python package.

not rated 4 7d ago A tokens not measured original Apache-2.0

NatiDvir/video-skills

Skill Claude CodeCodex

Write photorealistic live-action cinematic Seedance 2.0 prompts for Higgsfield, built on five grounding pillars that stop AI drift and floaty motion. Use for "cinematic film prompt", "shot like a movie", realistic body movement, grounded motion, restrained emotional close-ups, driving scenes, fight choreography…

not rated 4 +1 3mo ago A 111 tokens

分镜提示词

1014

wopanda/storyboard-prompt-skill

Skill Claude CodeCodex

A Chinese-language skill that turns a one-sentence video idea into a complete storyboard prompt after asking for missing details. It organises the prompt into basic settings, mood and image quality, and visual content for video models such as Seedance 2.0.

not rated 4 4mo ago A 87 tokens

prompt-architect

1015

hosanxiv/prompt-architect

Plugin Claude Code

Bundles 1 skill · 131 tokens together

Build high-quality, structured prompts using the 10-block framework from Anthropic's 'Prompting 101' workshop. Interview-style: asks 4 targeted questions, picks a task-type template, and produces a production-grade prompt with annotations and a test plan.

not rated 4 5mo ago A tokens not measured original MIT

minimax-worldbuilder

1016

Hearmeman24/minimax-worldbuilder

Skill Claude CodeCodex

Compose MiniMax-H3 video-with-native-audio prompts in H3's mandatory structured format — full-reference (Ref2VA) and text-to-video (T2VA) only. Use whenever the target model is MiniMax-H3: the user says MiniMax, MiniMax-H3, H3, ref2va / r2v / t2v, names a minimaxh3 checkpoint, or names the MiniMaxH3ReferenceToVideo /…

not rated 4 3mo ago A 333 tokens

qwen-mm-plugins-api

1017

ShuiHan268/dsh-qwen-mm-plugins

Skill Claude CodeCodex

Cloud MCP tools for understanding media, by model family. VL model: visionchat (caption/VQA), ocr, grounding (detect/locate objects). Omni model (reads frames + audio together): timestamped captioning, ASR (plain / controllable / multi-speaker diarized), temporal grounding, event counting, music captioning. Plus…

not rated 4 2mo ago A 110 tokens copy · 100% Apache-2.0

ky-ji/xie-saining-research-taste-skills

Skill Codex

Use when evaluating AI/ML research ideas, choosing research directions, framing problems, planning experiments, improving research taste or paper storytelling, resisting trend-following, or applying Saining Xie's interview-derived views on visual intelligence, world models, research culture, and scientific taste.

not rated 4 +3 5mo ago A 63 tokens original MIT

ahmedawan-oracle/claude-code-plugins

Plugin Claude Code

Bundles 14 skills · 1,388 tokens together

Operate the entire Oracle AI Data Platform (AIDP) Workbench in natural language — a 37-skill agent (not a single-engine orchestrator). Discovers your catalog into a grounding cache (FK/join hints + value dictionaries), turns plain English into accurate Spark SQL, and runs the full lakehouse SQL lifecycle…

not rated 4 +1 3mo ago A tokens not measured original MIT

hicham-bab/dbt-legacy-migration-skills

Skill Claude Code

Use when migrating Matillion pipelines/jobs to a dbt project — both Data Productivity Cloud pipelines (.orch.yaml / .tran.yaml) and Matillion ETL job exports (JSON). Maps each transformation component to a dbt model, snapshot, or macro; applies best-practice tests, docs, and contracts; validates the result against…

not rated 4 3d ago A 113 tokens original Apache-2.0

prompt-director

1021

adoin/video-generation-skills

Skill Claude CodeCodex

A guided tool for turning novels, scripts, scenes, or creative briefs into prompts for AI-generated images or videos.

not rated 4 +1 4mo ago A 62 tokens

unified-sf-docs-mcp

1022

TMTrevisan/unified-sf-docs-mcp

MCP server Claude CodeCodexCursor +2

Unified MCP server for scraping Salesforce developer documentation and searching a local SQLite vector-like database. Runs locally from the unified-sf-docs-mcp npm package.

not rated 4 8mo ago A tokens not measured

dsh-vision-mcp

1023

moton16/dsh-vision-mcp

MCP server Claude CodeCodexCursor +2

Give your text-only LLM eyes: an MCP server that reads images via external OpenAI-compatible vision models (OpenAI, Qwen-VL, GLM, Gemini, Grok, Ollama...) with automatic provider fallback on rate-limit/errors. Runs locally from the dsh-vision-mcp npm package.

not rated 4 2mo ago A tokens not measured original MIT

memograph

1025

Indhar01/MemoGraph

MCP server Claude CodeCodexCursor +2

A graph-based memory system for LLMs with intelligent retrieval using knowledge graphs, hybrid search, and semantic embeddings. Runs locally from the memograph Python package. Needs 7 environment variables to run.

not rated 4 2mo ago A tokens not measured original MIT

mcp-failure-lab

1026

anilloutombam/mcp-failure-lab

MCP server Claude CodeCodexCursor +2

A chaos-engineering and resilience-testing toolkit for Model Context Protocol servers. Runs locally from the mcp-failure-lab npm package.

not rated 4 2d ago A tokens not measured original MIT

inference-sh

1027

magentosh/superpowers

Plugin Claude Code

Bundles 86 skills · 9,995 tokens together

AI agent skills for models via inference.sh CLI - generate images with FLUX, create videos with Veo, call LLMs, search the web, and more.

not rated 4 3mo ago A tokens not measured fork

fgclip-mcp

1028

360CVGroup/FGCLIP-MCP

MCP server Claude CodeCodexCursor +2

MCP Server for FG-CLIP embedding services. Runs locally from the fgclip-mcp Python package.

not rated 4 1y ago A tokens not measured original Apache-2.0

smart-home-services

1029

balloob/llm-skills

Plugin Claude Code

Bundles 2 skills · 162 tokens together

Collection of smart home integration skills including Home Assistant automation and Ollama local AI capabilities.

not rated 4 9mo ago A tokens not measured

replicate

1030

sena-labs/replicate-mcp-server

MCP server Claude CodeCodexCursor +2

Universal MCP server giving any MCP client (Claude Desktop, claude.ai, Cursor, Cline, custom apps) native access to Replicate's full catalog: image, video, audio, music, speech, LLM, vision, upscale, inpaint, segment, transcription, embeddings, voice clon. Runs locally from the replicate-mcp-server npm package. Needs…

not rated 4 3d ago A Socket: warn tokens not measured original MIT

ods

1031

totonga/odsbox-jaquel-mcp

MCP server Claude CodeCodexCursor +2

ASAM ODS MCP Server - Query and analyze measurement data from ASAM ODS servers in AI agent systems. Runs locally from the odsbox_jaquel_mcp Python package. Needs 3 environment variables to run.

not rated 4 3d ago A Socket: warn tokens not measured

zilliz

1032

zilliztech/zilliz-plugin

Plugin Claude Code

Bundles 24 skills, 2 commands · 1,181 tokens together

Manage Zilliz Cloud clusters and Milvus vector databases from Claude Code using the zilliz-cli.

not rated 3 changed 14d ago A tokens not measured original Apache-2.0

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: