skills.rest
lucasmccomb
lucasmccombCOMMUNITY
Explore all skills by this author →
WebsiteGitHub

agent-native-audit

Audit codebases for agent-native architecture gaps and produce scored reports.

1|Updated Mar 19, 2026
Search Tags:#codebase#audit#subagent#composability#parity#agent-native#emergent-capability
One-click install
npx skills add https://github.com/lucasmccomb/ccgm --skill agent-native-audit-lucasmccomb
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: agent-native-audit
Source: https://github.com/lucasmccomb/ccgm/tree/main/modules/agent-native/skills/agent-native-audit
Command: npx skills add https://github.com/lucasmccomb/ccgm --skill agent-native-audit-lucasmccomb

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

The audit identifies whether a codebase exposes and structures agent-facing tools so that an AI assistant can achieve UI-driven outcomes reliably and safely.

Core Features & Use Cases

  • Parallel subagent analysis: launches eight concurrent reviewers (measure + critique per principle) to produce evidence-backed findings.
  • Concrete scoring and recommendations: returns counts, named violations, a 0–100 scorecard across parity, granularity, composability, and emergent capability, plus actionable first-PR fixes.
  • Use cases: pre-release readiness checks for agent-facing features, PR gating for tool-surface changes, and periodic audits for products marketed as agent-native.

Quick Start

Run the audit in full mode against the repository root to generate a scored report and write the run artifact to .agent-native-audit/.

Frequently Asked Questions about agent-native-audit

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

FAQPage Schema
What is an agent-native codebase audit and why do I need it before shipping features?▼

An agent-native codebase audit evaluates whether your repository exposes agent-facing tools so an AI assistant can achieve UI-driven outcomes reliably and safely. You need it to identify architecture gaps before releasing agent-facing features, ensuring pre-shipping readiness.

How do I score codebases for agent-native readiness and composability?▼

You score codebases for agent-native readiness by running parallel subagent analysis across parity, granularity, composability, and emergent capability. This generates a 0–100 scorecard with named violations and actionable first-PR fixes for your repository.

Can I run a headless codebase readiness check for periodic agent-native audits?▼

Yes, you can run a headless codebase readiness check by enabling optional headless JSON output. This allows periodic audits for products marketed as agent-native, generating scored reports without interactive UI elements.

Does the audit support PR gating for agent-facing tool-surface changes?▼

Yes, the audit supports PR gating for agent-facing tool-surface changes. It applies path-based inspection and parallel subagent measurement to produce evidence-backed findings, ensuring new routes or UI actions meet readiness requirements before merging.

What are the limitations of using subagent analysis for codebase readiness checks?▼

The subagent analysis is limited to evaluating product surfaces and repositories that expose agent-facing tools, routes, or UI actions. It does not assess general code quality outside the scope of agent-native architecture gaps and readiness scoring.

Related Skills

Explore More Software Engineering Skills→
opendatahub-ioopendatahub-io
36

shallow-clone

Creates a depth-1 clone of a Git repository in a temporary location.

Official
Intermediate
Miosa-osaMiosa-osa
226

code-review

Diff branch changes against a base and generate scored findings reports.

Official
Advanced
FFCfelps1FFCfelps1

ce-agent-native-architecture

Provide architectural patterns for building autonomous agent-native systems.

Community
Advanced
dotnetdotnet
209

client-compatibility

Select platform-appropriate agent spawning behavior for CLI, VS Code, and fallback surfaces.

Official
Advanced