MCP servers and skills: search, filter by where they run, and compare.
Nothing matches every word exactly, so these are the closest matches.
MCP server proof of concept: five user-scoped tools, token-forwarded authorization, and a local LLM chat demo
SkillMITFew starsNo key
Guides agents to interactively discover customer requirements for live, bidirectional multi-agent AI systems that process continuous streams of multimodal data for real-time technical guidance and saf
OfficialSkillApache-2.0★ 21k in collectionNo key
Guides users through discovering their database requirements, recommends a Google Cloud database based on a recommendation matrix, and assists in database creation. Use when a user asks 'What database
OfficialSkillApache-2.0★ 21k in collectionNo key
Terminal-first multi-agent runtime for building, testing and releasing games and apps with Claude Code, Codex and GigaChat.
SkillApache-2.0★ 1Few starsNo key
120 marketing skills as an AI marketing staff — plugin, portable skills, or an 8-bot team across 7 disciplines (narrative, SEO/GEO, social, email, paid, influencer, launch) on one contract, with 8 aud
SkillApache-2.0★ 3.0kNo key
Discipline layer for CLI agents. Run Claude Code, Codex, and Gemini CLI under one phase-gated runtime with deterministic detectors, replayable traces, persistent memory, and a chat REPL. MIT.
Node.jsMIT★ 2Few starsNo key
Technology-agnostic prompt generator that creates customizable AI prompts for scanning codebases and identifying high-quality code exemplars. Supports multiple programming languages (.NET, Java, JavaS
OfficialSkillMIT★ 40k in collectionNo key
A complete research system for neuroscience teams, inside Claude Code.
SkillMIT★ 7No key
AI Delivery Ops Skills Solution is a reusable Codex-style skills and templates for turning rough AI product ideas into phased delivery artifacts that teams can build, evaluate, govern, and review. The
SkillMIT★ 1Few starsNo key
CLUTCH — Claude Layered Unified Team Coordination Hub. Parallel PIV execution for Claude Code Agent Teams.
SkillMIT★ 4Few starsNo key
Node.jsMIT★ 5No key
Use when the user needs to bind users or teams to a Mobile Offline Profile so they actually receive offline sync on their devices. Without this, the profile exists in Dataverse but no one's app uses i
OfficialSkillMIT★ 1.0k in collectionNo key
Self-hosted autonomous-development BMAD framework platform — Telegram + console drive Claude Code through a typed, event-sourced spine. Python 3.12 · FastAPI · aiogram v3 · MCP stdio.
SkillMIT★ 2Few starsNo key
Mobile-first agentic coding harness: iPhone Telegram → Mac Mini → Claude Code + opencode + Hermes + Paperclip + Obsidian (Korean reference architecture)
SkillMIT★ 1Few starsNo key
Find your real patterns in your own data — chat exports, AI transcripts, app history — then change one thing and measure whether it moved. Most of it is safeguards.
SkillMITFew starsNo key
Drug pipeline competitive intelligence for pharmaceutical companies, biotech investors, and medical affairs teams starts here. This MCP server orchestrates **7 live data sources** — ClinicalTrials.gov
SkillMIT★ 1Few starsNo key
Claude Code skill that audits your CSS against your design system. Two agents scan in parallel — one hunts button and hover violations, one catches hardcoded colors, spacing drift, and typography issu
SkillMIT★ 3Few starsNo key
A Claude Code skill that assembles cross-functional AI teams around a problem statement and executes work in dependency-based waves
SkillMIT★ 3Few starsNo key
🖋️ OpenClaw AI Skill — 自我进化的三角色协作写作框架。Creator挖掘意图 → Executor产出多版本 → Reader六维评审 → 进化引擎越用越聪明。Self-evolving triadic writing with type-tagged memory matrix.
SkillMIT★ 3Few starsNo key
Claude Code skill: Turn feature ideas into implementation-ready plans with parallel research agents, agent teams, skill/MCP assignments, and Ralph Loop verification
SkillMIT★ 1Few starsNo key
A reusable Claude Code skill that decomposes long LLM-conversation exports into a navigable doc tree, imperative dev guide, and MADR-lite architecture decision records.
SkillMITFew starsNo key
Editorial pipeline for TechShu delivery teams - research, fact-check, draft, humanize, deliver. 22 skills, 13 agents, 10 quality gates. MIT licensed.
SkillMIT★ 1Few starsNo key
Deconstruct problems via First Principles. Classify hidden assumptions (Fact/Norm/Belief/Interest) and rank the "Most Dangerous" using a Vulnerability×Impact matrix. Rebuild conclusions from verified
SkillMIT★ 7No key
Cognithor · Agent OS: Local-first autonomous agent operating system. 19 LLM providers, 18 channels, 145 MCP tools, 6-tier memory, Agent Packs marketplace, zero telemetry. Python 3.12+, Apache 2.0.
PythonApache-2.0★ 157No key
Who stands behind it
