Start a fresh MCP setup
Pick this path when the team needs a working install route from zero, not a theory article.
Guides Hub
This hub exists for the moment after discovery. Use it when you already know the tool or stack you care about and now need setup steps, rollout guardrails, troubleshooting logic, or workflow decisions that hold up in production.
Coverage
9
implementation guides and operating docs live in this hub
Setup heavy
4
guides focused on first-run setup, MCP wiring, and environment bootstrap
Operator depth
4
workflow and troubleshooting pages for teams past simple installation
Start Here
Good implementation hubs do not dump every article into one flat list. They help you move from the current problem to the next correct page without opening five irrelevant tabs first.
Pick this path when the team needs a working install route from zero, not a theory article.
Use this lane when the install completed but skills, indexing, or runtime state still feel wrong.
Read this when the real decision is platform fit, workflow shape, or how the stack behaves in production.
Guide Categories
The fastest way to pick the right guide is to decide whether the real problem is setup, recovery, stack selection, or workflow design. Each section below groups articles by that outcome instead of by arbitrary publish order.
Use these when you are bootstrapping a new environment, integrating an MCP server, or validating first-run setup.
Install and bootstrap the NeonCtl MCP workflow with predictable local setup steps.
Good first stop when you need the broader Neon MCP rollout path.
Open guide →Use the latest NeonCtl bootstrap flow without stale package drift.
Best for teams validating the exact CLI init path and CI-safe setup checks.
Open guide →Step-by-step setup notes for MoltBook environments and MCP usage.
Use when a fresh workstation needs a clean MCP-ready baseline.
Open guide →Configure EvoMap as a real MCP service with validation checkpoints.
Focused on a concrete MCP server integration, not generic docs theory.
Open guide →Use these when installation looked successful but the runtime still behaves like the environment is half-configured.
Use these when you are choosing between stacks, not when you already know the implementation path.
Use these when the tools already exist and the real problem is repeatability, coordination, or monetization.
Production workflow patterns for repeatable orchestration and execution loops.
Helps operators turn one-off successes into repeatable process lanes.
Open guide →Keep prompts, memory, and runtime coordination aligned across agent sessions.
Designed for teams fighting prompt drift and memory fragmentation.
Open guide →A practical guide to monetization angles around OpenClaw-style ecosystems.
Useful when you need commercialization logic, not just implementation notes.
Open guide →Selection Matrix
Teams lose time when they read three setup articles to solve a recovery problem, or a workflow essay when they really needed a first-run command reference. This matrix keeps the hub practical.
| If your goal is… | Open this guide | Why this is the right next click |
|---|---|---|
| I need a fresh local setup that teammates can reproduce. | npx neonctl@latest init | It is the clearest route for validating first-run artifacts, prerequisites, and post-init checks. |
| I already know I need Neon MCP, but I want the broader rollout logic. | NeonCtl MCP init guide | It frames setup as a team baseline instead of a one-machine command. |
| The install finished, but skills are not showing up correctly. | MoltBook skills not found fix | It targets discovery and indexing breakpoints instead of generic install steps. |
| I need to compare two agent stacks before committing. | EvoMap vs OpenClaw | It covers positioning, workflow fit, and deployment tradeoffs in one place. |
| The tools are installed, but team execution is still inconsistent. | EvoMap workflow patterns | It focuses on repeatability and operating discipline rather than onboarding. |
Why this hub exists
AgentSkillsHub covers both discovery and implementation. Discovery pages are good when you are mapping the landscape. Guides matter later, when the team already has a tool or stack in mind and needs setup steps, file-side effects, rollout discipline, or a reliable comparison between two execution models.
That is why this hub is intentionally narrower than the full skills directory. The goal is not to list everything. The goal is to route operators to the right implementation depth fast enough that setup drift, false-positive success, and documentation sprawl do not slow down the actual work.
Also Browse
Claude Code plugins
Directory-style coverage for plugin discovery and installation paths.
Agent memory
Concept and tooling layer for stateful agent systems.
MCP servers
Browse server listings before choosing which implementation guide to open.
Datadog MCP Server
Datadog incidents, dashboards, and monitors inside Claude Code.
Perplexity MCP
Live search, research, and reasoning workflows for current web tasks.
Postman MCP Server
Hosted and local MCP setup for collections, environments, and API workflows.
Apify MCP Server
Actor discovery, tool selection, and hosted MCP setup for automation teams.
FAQ
The skills directory helps you discover tools and capabilities quickly. The guides hub is for implementation detail, rollout decisions, troubleshooting, and operating patterns after discovery.
Start with the npx neonctl@latest init guide if you need the exact CLI setup path. Move to the NeonCtl MCP init guide when you need the broader rollout policy and team baseline.
No. The hub is organized by outcome: setup, troubleshooting, comparison, and workflow. Most guides call out where editor-specific or runtime-specific behavior matters.
Use the start-here paths first, open one guide that matches the current bottleneck, and only add a second guide when the problem changes from setup to troubleshooting or from implementation to operating model.
Return to the skills directory when you are exploring tooling options. Stay in the guides hub when you already know the tool and need steps, guardrails, validation, or workflow decisions.