Can I add custom keys to the findings table?

No, this skill enforces a strict contract of 11 specific keys (including finding_id, severity, and dimension) to ensure compatibility across all audit-fleet lanes.

What is the primary purpose of the Audit Fleet Output Contract?

It ensures all Claude Code audit outputs follow a rigid, pre-defined structure to allow for automated processing and merging of reports without manual formatting.

How are severity levels handled?

The skill enforces a strict enum consisting of critical, warning, or info, preventing ambiguous or inconsistent labeling across different audit sessions.

What sections are required in the audit output?

Reports must include an Executive Summary, Findings, Quick Wins, and High-Impact Expansions in that exact order.

Audit Fleet Output Contract

Name: Audit Fleet Output Contract
Author: nsalvacao

bynsalvacao

0•

Security & Testing

Enforces a standardized, deterministic output format for audit reports to ensure seamless data merging and consistency across automated audit lanes.

The Audit Fleet Output Contract skill provides a rigorous structural framework for Claude Code during auditing tasks, ensuring that all findings and reports follow a canonical schema. By enforcing a strict markdown table structure and specific metadata keys—such as severity, impact, and effort—it eliminates the manual rework typically required when merging outputs from multiple automated audit processes. This skill is essential for teams running parallel audit tracks that require a unified source of truth and deterministic reporting for security, technical debt, or architectural reviews.

Key Features

01Validates findings against 11 mandatory keys including impact, effort, and confidence.

02Standardizes severity levels into a strict critical, warning, and info enum.

03Categorizes recommendations into actionable Quick Wins versus long-term High-Impact Expansions.

04Automatically formats finding data into structured markdown tables for easy parsing.

05Enforces a canonical four-section report structure from Executive Summary to High-Impact Expansions.

060 GitHub stars

Use Cases

01Consolidating security audit results from multiple parallel automated scanning lanes.

02Standardizing technical debt reporting across distributed engineering teams.

03Generating merge-ready audit reports for continuous integration and delivery pipelines.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add nsalvacao/nsalvacao-claude-code-plugins audit-fleet-output-contract

For use in Claude.ai and ChatGPT

Key Features

01Validates findings against 11 mandatory keys including impact, effort, and confidence.

02Standardizes severity levels into a strict critical, warning, and info enum.

03Categorizes recommendations into actionable Quick Wins versus long-term High-Impact Expansions.

04Automatically formats finding data into structured markdown tables for easy parsing.

05Enforces a canonical four-section report structure from Executive Summary to High-Impact Expansions.

060 GitHub stars

Use Cases

01Consolidating security audit results from multiple parallel automated scanning lanes.

02Standardizing technical debt reporting across distributed engineering teams.

03Generating merge-ready audit reports for continuous integration and delivery pipelines.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add nsalvacao/nsalvacao-claude-code-plugins audit-fleet-output-contract

For use in Claude.ai and ChatGPT