Claude Code Plugins

Community-maintained marketplace

Feedback

runtime-diagnostics

@Cloudhabil/AGI-Server
0
0

Run local system diagnostics and model benchmarks across CPU, memory, GPU, NPU, Docker, and Kubernetes; use when asked to benchmark, health-check, or collect environment diagnostics, and store a summary in memory.

Install Skill

1Download skill
2Enable skills in Claude

Open claude.ai/settings/capabilities and find the "Skills" section

3Upload to Claude

Click "Upload skill" and select the downloaded ZIP file

Note: Please verify skill by going through its instructions before using it.

SKILL.md

name runtime-diagnostics
description Run local system diagnostics and model benchmarks across CPU, memory, GPU, NPU, Docker, and Kubernetes; use when asked to benchmark, health-check, or collect environment diagnostics, and store a summary in memory.

Runtime Diagnostics

Use this skill to collect a consistent diagnostics snapshot with timeouts and safe fallbacks.

Workflow

  1. Run the diagnostics script:
python skills/automation/runtime-diagnostics/scripts/diagnostics.py --runs 1
  1. If Docker or CIM access fails, rerun with elevated permissions.
  2. Share the JSON output summary and note any failures/timeouts.
  3. The script stores a semantic memory entry with the full context.

Notes

  • The script times out Kubernetes queries; it will record errors instead of hanging.
  • If sentence-transformers or embedding backends are missing, the benchmark can still run.
  • Use --runs 3 for a deeper model benchmark (slower).