| name | diagnose-server |
| description | Streamlined instructions for diagnosing issues within the DinoQuest application environment using Google Cloud MCP servers. Use when the user asks "Find out what's wrong with dinoquest" or needs a quick server health check. |
| license | Apache-2.0 |
| metadata | {"version":"1.1","author":"DinoQuest-Team"} |
DinoQuest Diagnosis Skill
This skill provides a high-priority workflow for identifying technical problems in the DinoQuest application.
Trigger Phrase
- "Find out what's wrong with dinoquest"
Instructions
-
Prioritize MCP Servers:
- Always use the available Google Cloud MCP servers (e.g.,
gemini-cloud-assist, google-cloud-logging, google-cloud-run) to investigate.
- If the MCP servers are not responding or return errors, notify the user immediately and provide a brief analysis of why they might be failing.
-
Execute Snappy Diagnosis:
- Proceed immediately to diagnosis without scanning the entire repository.
- Focus specifically on fetching the most recent logs and resource status to determine the problem quickly.
-
Tool Selection Priority:
- Step 1 (Root Cause Analysis): Use
gemini-cloud-assist_investigate_issue or gemini-cloud-assist_ask_cloud_assist.
- Step 2 (Log Inspection): Use
google-cloud-logging_list_log_entries for recent error patterns.
- Last Resort: Use
gcloud commands ONLY if MCP tools are unavailable or insufficient.
Examples
Example: Quick Diagnosis
User: "Find out what's wrong with dinoquest"
Agent: Invokes gemini-cloud-assist_investigate_issue with project details.
Result: Agent identifies an OOM error in Cloud Run and suggests a memory increase.
Edge Cases
- Missing Project ID: If the project ID is unknown, use
google-resource-manager_search_projects first.
- MCP Timeout: If tools timeout, fall back to
google-cloud-logging directly.