integrated-audit

Synthesize findings from multiple analytical audits into an integrated structural diagnosis of a written work.

Updated Sep 8, 2026
One-click install
npx skills add https://github.com/jkutianski/Rizome-and-AI --skill integrated-audit-jkutianski
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: integrated-audit
Source: https://github.com/jkutianski/Rizome-and-AI/tree/main/.agents/skills/integrated-audit
Command: npx skills add https://github.com/jkutianski/Rizome-and-AI --skill integrated-audit-jkutianski

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When multiple analytical audits produce dozens of findings about a manuscript, it becomes hard to tell which issues are real, which overlap, which are architectural versus local, and which apparent problems should actually be preserved. This Skill consolidates those findings into one coherent diagnosis of the work as a whole. ## Core Features & Use Cases - Finding Synthesis: Groups convergent findings from different audits under shared root causes, separates independent issues, and prevents double-counting of the same underlying problem. - Severity and Scope Classification: Distinguishes local issues from architectural ones and grades severity from critical to non-problematic, while protecting productive tensions, asymmetry, and authorial voice. - Structured Final Report: Produces a mandatory diagnostic report covering architecture, real problems, nodes, connections, propagations, contradictions, preservation decisions, prioritized interventions, and a final readiness verdict. - Use Case: After running several specialized audits (semantic precision, causality, redundancy, philosophical risk) on a long-form essay, use this Skill to merge their outputs into a single verdict on whether the work is ready for final editing. ## Quick Start Synthesize the findings from all previous audit rounds of my manuscript into an integrated diagnosis with a final readiness verdict.

Frequently Asked Questions about integrated-audit

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

FAQPage Schema
How do I combine results from multiple manuscript audits?

Collect findings by domain, identify convergent findings that share one root cause, and keep independent findings separate. Then classify each as local or architectural, assign severity, and produce an integrated diagnosis rather than concatenating individual audit outputs.

What is the difference between a local and an architectural problem in a text?

A problem is local if correcting it leaves the overall structure unchanged. It is architectural if fixing it requires changing node boundaries, domain organization, conceptual definitions, major relations, trajectory, or the ending of the work.

How do I avoid double-counting findings from different review passes?

Identify the shared root problem behind overlapping findings, retain the strongest evidence, and record the rest as confirmations. Convergence increases confidence in the finding but does not multiply its severity.

When should a contradiction in a manuscript be preserved instead of fixed?

Preserve it when the tension is productive, meaning it generates analytical work, such as map versus territory or stability versus transformation. Only genuine, semantic, or structural contradictions that damage coherence require intervention.

What are the limitations of an integrated audit?

It does not rewrite the work, rerun individual audits, or force consensus between conflicting findings. It is a diagnostic synthesis, so it can end with unresolved questions when evidence is insufficient rather than inventing resolutions.