A specialized skill for installing and utilizing Switchyard's read-only MCP runtime diagnostics surface for precise AI agent debugging.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install switchyard-runtime-diagnostics
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install switchyard-runtime-diagnostics using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is Switchyard Runtime Diagnostics?
Switchyard Runtime Diagnostics is a technical bridge that empowers AI agents to inspect and understand their operational environment without the risk of unauthorized state changes. By integrating this module via Openclaw Skills, developers can grant agents the ability to connect to a local stdio MCP server, allowing for deep introspection of provider boundaries and runtime truth.
This skill is centered on grounding agent claims in reality, ensuring that internal and external blockers are clearly identified. It provides a standardized way to handle partial and read-only data states, making it an essential component for maintainable and observable AI workflows.
Switchyard Runtime Diagnostics Use Cases
- Diagnosing specific provider or runtime boundary issues within an AI system.
- Empowering agents to verify catalog truth before initiating human-requested actions.
- Separating internal software blockers from external API or environment blockers.
- Running automated diagnostic loops to verify host configurations.
How Switchyard Runtime Diagnostics Works
- The agent identifies the need for runtime truth and triggers the Openclaw Skills diagnostic workflow.
- The host environment initializes the local stdio MCP server based on the provided configuration files.
- The agent attaches to the Switchyard MCP server to gain a read-only view of the runtime and catalog surface.
- Diagnostics are run to evaluate auth-status and package-readiness.
- The agent reports findings, strictly distinguishing between supported and partial features without modifying external accounts.
Switchyard Runtime Diagnostics Setup
To get started with Switchyard Runtime Diagnostics, follow these installation steps:
- Review the installation prerequisites in the references directory.
- Load the appropriate host configuration for your environment:
# For Openclaw environments
cat references/OPENCLAW_MCP_CONFIG.json
# For OpenHands environments
cat references/OPENHANDS_MCP_CONFIG.json
- Initialize the diagnostic loop to verify connectivity:
# Execute the demo diagnostic path
./scripts/run_demo_diagnostic.sh
Switchyard Runtime Diagnostics Data Schema & Taxonomy
The skill organizes its diagnostic data and metadata according to the following taxonomy:
| Data Component | Description | Access Level |
|---|---|---|
| Runtime Truth | Live state of the MCP server boundaries | Read-Only |
| Catalog Surface | List of available tools and capabilities | Read-Only |
| Auth Status | Current authentication readiness level | Read-Only |
| Blockers | Classification of internal vs external issues | Metadata |
All generated diagnostic logs are grounded in partial, read-only, and package-ready states.
Switchyard Runtime Diagnostics Advanced Features
- Support for multiple host configurations including OpenHands and Openclaw Skills native formats.
- Automated diagnostic loops for rapid proof-of-path verification.
- Granular guardrails preventing unauthorized writes to external accounts.
- Intelligent status mapping that prevents 'partial' support from being misreported as 'full' support.
SKILL.md
Loading
Related Openclaw Skills
Shopflow Read-only Packet
A technical integration packet designed to teach AI agents how to interface with Shopflow as a read-only surface using MCP servers.
Polymarket Value Trader
A disciplined prediction market trading skill for AI agents that utilizes value-based analysis to trade on Simmer and Polymarket venues.
Memory Taxonomist
A specialized skill for AI agents to transform raw, unstructured information into a disciplined taxonomy of durable knowledge.
Dual-Mode Reasoner
A disciplined reasoning protocol that protects against high-stakes errors by intelligently switching between rapid execution and deep, risk-aware deliberation.
Movi Review-First Bundle
A safety-oriented skill for AI agents to manage local MCP servers with a review-first approach to data mutation and batch analysis.
Article Taster
A multi-dimensional article evaluator that filters low-quality content and detects AI-generated text to save you valuable reading time.
Featured*
