Browse how real teams set up AI coding tools
mdgrok tracks public repos and pulls out the AGENTS.md, CLAUDE.md, SKILL.md, and MCP config patterns teams actually commit.
Categories
Browse the ecosystem
The skill catalog starts broad so people can browse by intent, then narrows to task-shaped skills once they know the job they want done.
Design & UX
4,934 items
Interface design, critique, copy, polish, and front-end presentation work.
Build & Generate
1,552 items
Implementation-first skills that create starter code, scaffolds, and working drafts.
Review & Improve
5,244 items
Code review, cleanup, audits, and quality passes over work that already exists.
Test & Verify
5,179 items
QA, assertions, browser checks, and repeatable proof that something works.
Research & Reference
13,150 items
API lookup, repo exploration, and reverse engineering when you need understanding first.
Data & Observability
6,110 items
Logs, dashboards, SQL, metrics, and investigation workflows driven by data.
Delivery & Ops
4,614 items
Deploys, incidents, infrastructure work, and production-facing operating procedures.
Team Workflows
1,381 items
Approvals, triage, release coordination, and repeatable team automation.
Other
9,597 items
Skills that do not fit a narrower decision bucket yet.
Top Pages
Open strong examples first
Start with the pages people use most: strong repo setups, high-scoring skills, and the MCP servers that show up across the index.
AGENTS.md / CLAUDE.md
Top repository setups
DawVert/DawVert
AGENTS.md CLAUDE.mdPython·385 stars
DawVert - The DAW ConVERTer
GthiN89/FuckingCheapFlipperZero-DIY-Flipper-zero-The-real-on
AGENTS.mdC·157 stars
DIY suitable forced open source hw Flipper zero + Momentum optimized to run on it. Off the shelf modules, arduino skill level needed only, price from 10 (minimal) to 30 (complete) euros per build
strvmarv/total-recall
AGENTS.md0 stars
Multi-tiered memory and knowledge base plugin for TUI coding assistants (Claude Code, Copilot CLI, OpenCode, Cline, Cursor)
SKILL.md
Top reusable workflows
concept-diagrams
Build & Generate Generate Code ScaffoldingNousResearch/hermes-agent
Generate flat, minimal light/dark-aware SVG diagrams as standalone HTML files, using a unified educational visual language with 9 semantic color ramps, sentence-case typography, and automatic dark mode. Best suited for educational and non-software visuals — physics setups, chemistry mechanisms, math curves, physical objects (aircraft, turbines, smartphones, mechanical watches), anatomy, floor plans, cross-sections, narrative journeys (lifecycle of X, process of Y), hub-spoke system integrations (smart city, IoT), and exploded layer views. If a more specialized skill exists for the subject (dedicated software/cloud architecture, hand-drawn sketches, animated explainers, etc.), prefer that — otherwise this skill can also serve as a general-purpose SVG diagram fallback with a clean educational look. Ships with 15 example diagrams.
Generates starter code, templates, and project structure.
component-fixtures
Design & UX Build UImicrosoft/vscode
Use when creating or updating component fixtures for screenshot testing, or when designing UI components to be fixture-friendly. Covers fixture file structure, theming, service setup, CSS scoping, async rendering, and common pitfalls.
Creates new interfaces or redesigns in code.
skill-writing
Test & Verify Verify Product Behaviormillionco/expect
Write and improve agent skills (SKILL.md files). Use when creating new skills, refactoring existing ones, debugging why an agent ignores instructions, or improving compliance. Covers prompt structure, TDD for skills, description optimization, and rationalization bulletproofing.
Tests flows, asserts behavior, and gathers evidence that something works.
MCP
Most-used servers
nrepl
nrepl-mcp-server
Open the catalog page to inspect adoption, sample configs, and the config-derived security signal.
jira
@mcp-devtools/jira
Open the catalog page to inspect adoption, sample configs, and the config-derived security signal.
responsible-vibe-mcp
@codemcp/workflows@latest
Open the catalog page to inspect adoption, sample configs, and the config-derived security signal.
The Files
What are these files?
mdgrok separates the file types because they do different jobs. The browse experience is unified, but the scoring stays honest about what each artifact does.
AGENTS.md
Project instructions meant to work across coding assistants. mdgrok gives them a 0-100 quality score and a separate security result.
CLAUDE.md
Claude Code memory and repo instructions. These use the same quality scale as AGENTS.md, with the same security scanner alongside it.
SKILL.md
Task-shaped instructions and reusable workflows. mdgrok evaluates quality, browsing metadata, and security findings without treating them like repo-wide operating guides.
MCP configs
Server references found inside repos. mdgrok extracts package refs, commands, env requirements, and transport, then runs the shared security scanner over the config content.