Claude Skill

dcc-diagnostics

Infrastructure skill — DCC-agnostic observability primitives: capture screenshots, query audit logs, inspect tool performance metrics, and monitor process health. Works in any DCC environment (Maya, Blender, Houdini, Unreal, etc.) or standalone Python. Use for debugging any skill

LLM Mart · 0 points · 0 views 3 listing impressions 0 install-command copies

#observability

Download dcc-mcp-dcc-mcp-core-examples_skills_dcc-diagnostics-8d8c969.zip · 8 KB

Install

skills CLI npx skills add https://github.com/dcc-mcp/dcc-mcp-core/tree/main/examples/skills/dcc-diagnostics
Claude Code claude plugin marketplace add https://llmmart.ai/marketplace.json && claude plugin install dcc-mcp-dcc-mcp-core@llmmart
Git git clone https://github.com/dcc-mcp/dcc-mcp-core.git

The skills CLI installs just this skill, for any of its supported agents. Claude Code installs the whole dcc-mcp/dcc-mcp-core collection as a plugin from our marketplace. Git is the plain clone.

Skill manifest

DCC Diagnostics

Cross-DCC observability and debugging tools powered by dcc-mcp-core.

All tools work in any DCC environment (Maya, Blender, Houdini, Unreal, 3ds Max) or standalone Python — no DCC-specific APIs required.

Tools

dcc_diagnostics__screenshot

Capture the current screen or a specific window as a PNG/JPEG image. Backed by the dcc_mcp_core.Capturer class which uses:

  • Windows: DXGI Desktop Duplication API (<16ms per frame)
  • Linux: X11 XShmGetImage
  • Fallback: Mock synthetic backend (headless/CI)

dcc_diagnostics__audit_log

Query the sandbox audit log from dcc_mcp_core.SandboxContext. Returns recent tool invocations with outcome (success/denied) and timestamps.

dcc_diagnostics__tool_metrics

Inspect per-tool performance counters from dcc_mcp_core.ToolRecorder: invocation count, success rate, average latency, P95/P99 percentiles.

dcc_diagnostics__process_status

Check process health via dcc_mcp_core.PyProcessMonitor. Lists tracked PIDs and their liveness status.

Usage with any DCC MCP server

import os

os.environ["DCC_MCP_SKILL_PATHS"] = "/path/to/dcc-diagnostics"

from dcc_mcp_maya import start_server  # or dcc_mcp_blender, etc.

handle = start_server(port=8765)
# dcc_diagnostics__screenshot is now available as an MCP tool
Files (dcc-mcp-core)
  • scripts
    • audit_log.py 5.6 KB
      Loading…
    • process_status.py 3.4 KB
      Loading…
    • screenshot.py 3.1 KB
      Loading…
    • tool_metrics.py 5.3 KB
      Loading…
  • SKILL.md 2.2 KB
    ---
    name: dcc-diagnostics
    description: >-
      Infrastructure skill — DCC-agnostic observability primitives: capture
      screenshots, query audit logs, inspect tool performance metrics, and monitor
      process health. Works in any DCC environment (Maya, Blender, Houdini, Unreal,
      etc.) or standalone Python. Use for debugging any skill failure or verifying
      DCC state. Not for primary task execution — use a domain skill for actual DCC
      operations.
    license: MIT
    metadata:
      dcc-mcp:
        dcc: python
        recall-context:
          app_type: python
          domain: diagnostics
          workflow_stage: diagnostics
          task_category: diagnose
        version: "1.0.0"
        layer: infrastructure
        search-hint: "screenshot, capture, audit log, metrics, performance, process monitor, diagnostics, debug, health check, observability"
        tags: "diagnostics, observability, screenshot, audit, metrics, debug, infrastructure"
        tools: tools.yaml
    ---
    
    # DCC Diagnostics
    
    Cross-DCC observability and debugging tools powered by `dcc-mcp-core`.
    
    All tools work in any DCC environment (Maya, Blender, Houdini, Unreal, 3ds Max)
    or standalone Python — no DCC-specific APIs required.
    
    ## Tools
    
    ### `dcc_diagnostics__screenshot`
    
    Capture the current screen or a specific window as a PNG/JPEG image.
    Backed by the `dcc_mcp_core.Capturer` class which uses:
    
    - **Windows**: DXGI Desktop Duplication API (<16ms per frame)
    - **Linux**: X11 XShmGetImage
    - **Fallback**: Mock synthetic backend (headless/CI)
    
    ### `dcc_diagnostics__audit_log`
    
    Query the sandbox audit log from `dcc_mcp_core.SandboxContext`.
    Returns recent tool invocations with outcome (success/denied) and timestamps.
    
    ### `dcc_diagnostics__tool_metrics`
    
    Inspect per-tool performance counters from `dcc_mcp_core.ToolRecorder`:
    invocation count, success rate, average latency, P95/P99 percentiles.
    
    ### `dcc_diagnostics__process_status`
    
    Check process health via `dcc_mcp_core.PyProcessMonitor`.
    Lists tracked PIDs and their liveness status.
    
    ## Usage with any DCC MCP server
    
    ```python
    import os
    
    os.environ["DCC_MCP_SKILL_PATHS"] = "/path/to/dcc-diagnostics"
    
    from dcc_mcp_maya import start_server  # or dcc_mcp_blender, etc.
    
    handle = start_server(port=8765)
    # dcc_diagnostics__screenshot is now available as an MCP tool
    ```
    
  • tools.yaml 3.7 KB
    Loading…

Comments (0)

Sign in to join the conversation.

No comments yet.

Reviews (0)

No reviews yet.