cross-cut-analysis

Scan boundary.yaml files to identify cross-cutting coordination risks and emit YAML reports.

Updated Aug 23, 2026
One-click install
npx skills add https://github.com/ahlerjam/mmp --skill cross-cut-analysis
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cross-cut-analysis
Source: https://github.com/ahlerjam/mmp/tree/main/skills/cross-cut-analysis
Command: npx skills add https://github.com/ahlerjam/mmp --skill cross-cut-analysis

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Cross-chunk parallel work can fail or corrupt shared state when multiple chunks overlap files, implicitly depend on each other, or touch the same domains and resources without explicit coordination.

Core Features & Use Cases

  • Boundary-driven risk analysis: Scans all chunk boundary.yaml files (allow, produces, consumes, touches_domains, tooling_deps, depends_on) to find coordination gaps.
  • Five risk dimensions: Reports file-overlap conflicts, domain-tag conflicts, implicit produce/consume dependencies, concurrency hazards on shared state channels, and shared tooling requirements.
  • Actionable structured output: Produces a YAML cross_cuts report with severity and recommendations, plus an audit event for traceability.

Quick Start

Run /mmp:cross-cut-analysis to generate a YAML cross-cut report for the current set of chunks and see which findings require serialization before dispatch.

Frequently Asked Questions about cross-cut-analysis

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

FAQPage Schema
How do I detect cross-chunk coordination risks before parallel dispatch?

Concurrency hazards on shared state channels are identified by analyzing boundary.yaml metadata across chunks, specifically examining produces, consumes, and touches_domains fields. The resulting cross-cut report flags these concurrency risks with appropriate severity and recommends serialization fixes.

What is cross-cut analysis and when do I need it for orchestrator dispatch?

Cross-cut analysis identifies five risk dimensions: file-overlap conflicts, domain-tag conflicts, implicit produce/consume dependencies, concurrency hazards on shared state channels, and shared tooling requirements. It scans all chunk boundary.yaml files and outputs a structured YAML report with severity and scheduling recommendations.

How do I generate a cross-cut analysis report for chunks with boundary.yaml metadata?

Run the cross-cut-analysis command to scan all chunk boundary.yaml files and generate a YAML report. The report details file-overlap conflicts, domain-tag conflicts, implicit dependencies, concurrency hazards, and shared tooling requirements, while automatically recording a cross_cut_analysis_run audit event.

Why does parallel work fail when chunks touch the same domains and resources?

Parallel work fails due to uncoordinated overlaps in files, domains, or shared state channels. Cross-cut analysis detects these issues by scanning boundary.yaml metadata and producing a structured YAML report that recommends scheduling fixes to serialize conflicting chunks before dispatch.

Can I run cross-cut analysis manually when parallel dispatch waves fail?

Cross-cut analysis serves as a manual diagnostic tool when parallel dispatch waves fail. It scans boundary.yaml files to identify file-overlap conflicts, domain-tag conflicts, and concurrency hazards, producing a YAML report with severity levels and a cross_cut_analysis_run audit event for traceability.

What dependencies do I need to perform a cross-cut analysis on chunks?

No external dependencies are required. The analysis solely requires existing .orchestrator/chunks/*/boundary.yaml files containing metadata fields like produces, consumes, touches_domains, tooling_deps, and depends_on to detect coordination risks and output YAML findings.