max-critic

Identify quality blockers in MAX patches using deterministic Python critic modules.

41|1|Updated Mar 12, 2026
One-click install
npx skills add https://github.com/taylorbrook/MAX-MSP_CC_Framework --skill max-critic
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: max-critic
Source: https://github.com/taylorbrook/MAX-MSP_CC_Framework/tree/main/.claude/skills/max-critic
Command: npx skills add https://github.com/taylorbrook/MAX-MSP_CC_Framework --skill max-critic

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Orchestrates an automated, deterministic quality-assurance loop for generated MAX patches, ensuring blockers are surfaced and resolved before delivery.

Core Features & Use Cases

  • Orchestrates the generate-review-revise cycle using Python critic modules to analyze patch structure and signal flow.
  • Invokes and aggregates results from review_patch() and related critics to produce actionable revision requests.
  • Tracks revision history, annotates warnings/notes inline, and handles optional RNBO, package, and external critiques.
  • Enforces a soft limit after three rounds and escalates identical unresolved findings after five rounds.
  • Supports references/critic-protocol docs to define the loop workflow and data schemas.

Quick Start

Run the critic on the generated patch to identify blockers, then apply revisions or annotations before finalizing the patch.

Frequently Asked Questions about max-critic

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

FAQPage Schema
How do I automate quality critique and revision for generated MAX patches?

Automated MAX patch critique applies deterministic Python critic modules to analyze patch structure and signal flow, surfacing quality blockers across iterative revisions to enforce package and compliance conventions before delivery.

What happens when unresolved blockers persist through multiple patch revision rounds?

Patch revision loops enforce a soft-limit pause after three rounds of unresolved blockers, then escalate identical findings after five rounds, ensuring persistent structural or signal flow issues are flagged for manual review.

Can I use deterministic critics to validate RNBO and external package compliance in MAX patches?

Max patch critique supports optional RNBO, package, and external critiques, aggregating results from review_patch() to validate signal flow correctness and package compliance conventions across iterative revision cycles.

What's the best way to structure a generate-review-revise cycle for MAX patch quality assurance?

The generate-review-revise cycle invokes Python critic modules on patch_dict and optional code_context, tracking revision history and annotating inline warnings to produce actionable revision requests for structural compliance.

Do I need specific data schemas to run automated patch quality checks?

Automated patch quality checks require a patch_dict input and optional code_context, utilizing defined critic-protocol data schemas and workflow references to execute deterministic structure and signal flow analysis.

Why does my MAX patch critique loop keep pausing before finalizing the output?

Patch critique loops trigger soft-limit pauses after three revision rounds to prevent infinite cycling, escalating only when identical unresolved findings persist after five rounds of analysis.