meta-analyst

Analyze AutomationOne debug reports to detect contradictions and cross-layer cascades.

Updated Aug 27, 2026
One-click install
npx skills add https://github.com/Auto-one-Family/Automation-One --skill meta-analyst
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: meta-analyst
Source: https://github.com/Auto-one-Family/Automation-One/tree/main/.claude/skills/meta-analyst
Command: npx skills add https://github.com/Auto-one-Family/Automation-One --skill meta-analyst

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Identifies and documents cross-report contradictions, timing correlations, and cascading failures across AutomationOne debug reports to provide a single, sourced overview for the Technical Manager without proposing solutions.

Core Features & Use Cases

  • Consolidation: Reads all agent reports in .claude/reports/current/ and inventories available findings and timestamps.
  • Cross-layer Correlation: Matches ESP32, MQTT, server, frontend, and DB events on a timeline to reveal propagation chains and likely root causes.
  • Discrepancy Detection & Prioritization: Flags conflicting statements across reports, notes missing report gaps, and produces a prioritized, source-linked META_ANALYSIS.md for triage.
  • Use Case: After a failing test session, run the meta-analyst to produce a chronological, source-referenced summary of critical cascades and contradictions for the Technical Manager to delegate follow-up.

Quick Start

Ask the meta-analyst to analyze all Markdown reports in .claude/reports/current/ for session <session_id> and produce .claude/reports/current/META_ANALYSIS.md with sourced findings.

Frequently Asked Questions about meta-analyst

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

FAQPage Schema
How do I correlate ESP32, MQTT, and server logs for cross-report debug analysis?

Cross-report debug analysis matches ESP32, MQTT, server, frontend, and DB events on a timeline to reveal propagation chains. It reads Markdown reports from .claude/reports/current/ and extracts timestamps to detect cross-layer cascades and likely root causes.

What is the best way to detect contradictions across multiple agent debug reports?

Detecting contradictions across agent debug reports involves reading all Markdown files in .claude/reports/current/ and comparing severity markers and timestamps. The process flags conflicting statements across ESP32, MQTT, server, frontend, and DB reports to identify discrepancies.

How do I generate a prioritized META_ANALYSIS.md from post-test debug reports?

To generate a prioritized META_ANALYSIS.md, the tool inventories all findings and timestamps from session reports in .claude/reports/current/. It then produces a source-linked summary of critical cascades and contradictions for triage without proposing solutions.

Can I use cross-report correlation to find root causes for cascading failures in ESP32 and MQTT systems?

Cross-report correlation finds root causes for cascading failures by matching ESP32 and MQTT events on a chronological timeline. This reveals propagation chains across server, frontend, and database layers during debugging and post-test reviews.

Does the meta-analyst require reports to be in a specific format for incident triage?

Incident triage requires reports to be Markdown files located in .claude/reports/current/. The analysis reads these files, extracts severity markers and timestamps, and notes missing report gaps to produce a sourced overview.

What are the limitations of cross-report analysis for AutomationOne debug sessions?

Cross-report analysis limitations include providing a sourced overview without proposing solutions. It relies entirely on reading existing Markdown reports in .claude/reports/current/ and cannot generate findings if agent reports are missing or lack severity markers.