dogfood

Diagnose Exarchos MCP tool failures and classify root causes into code, documentation, or user errors.

33|Updated Jan 5, 2026
One-click install
npx skills add https://github.com/lvlup-sw/exarchos --skill dogfood-lvlup-sw
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dogfood
Source: https://github.com/lvlup-sw/exarchos/tree/main/skills-src/dogfood
Command: npx skills add https://github.com/lvlup-sw/exarchos --skill dogfood-lvlup-sw

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill performs retrospective triage of failed Exarchos MCP tool calls to determine why workflows and tool invocations failed, preventing wasted effort and guiding corrective action across code, documentation, or user behavior.

Core Features & Use Cases

  • Authoritative Debug Trace: Run MCP self-service queries (describe, event queries, runbooks, and views) to build a ground-truth diagnostic picture independent of conversation history.
  • Session Scan & Correlation: Supplement server-side traces with session tool-call review to map errors to attempted actions and parameters for reproducible evidence.
  • Root-Cause Classification: Classify each failure into exactly one bucket — code bug, documentation issue, or user error — with trace-backed rationale and severity.
  • Actionable Reporting: Produce a structured report following the included template, draft GitHub issue bodies for code/docs findings, and offer to file them with user confirmation.
  • Use Case: Run at the end of a workflow session or when encountering INVALID_INPUT, VALIDATION_ERROR, sequence conflicts, or repeated retries to quickly identify whether to file a bug, update docs, or retrain an agent.

Quick Start

Invoke the dogfood skill or run the /dogfood command scoped to the workflow to generate a trace-backed triage report and draft issue suggestions.

Frequently Asked Questions about dogfood

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I diagnose Exarchos MCP tool failures and determine the root cause?

To diagnose Exarchos MCP tool failures, run MCP self-service queries and session tool-call scans to categorize each failure as a code bug, documentation issue, or user error with trace-backed rationale.

When should I run a triage playbook for failed MCP workflow runs?

Run a triage playbook for failed MCP workflow runs when encountering INVALID_INPUT, VALIDATION_ERROR, sequence conflicts, or repeated retries to identify whether to file a bug, update docs, or retrain an agent.

What is the best way to categorize MCP tool invocation errors into code bugs or user errors?

The best way to categorize MCP tool invocation errors is using event stream analysis and runbook comparisons to build a ground-truth diagnostic picture, classifying each failure into exactly one bucket with severity.

Does this debugging process support all Exarchos tools or only specific workflow calls?

This debugging process applies exclusively to Exarchos tools exarchos_workflow, exarchos_event, exarchos_orchestrate, exarchos_view, and exarchos_sync when triaging workflow runs and session tool calls.

How do I generate GitHub issue drafts from MCP event log analysis?

Generate GitHub issue drafts from MCP event log analysis by producing a structured trace-backed report following the included template, then offering to file the drafted issue bodies for code or documentation findings with user confirmation.