hello auditkit

Security Community

Skill notes

Table of Contents

Entry Point Overview Core Principles Audit Execution Reference Files

Entry Point

On skill invocation, first determine the audit target: User Input · Action No target specified · Show welcome message and usage guide (see below) File path provided · Audit the specified file Directory path provided · Scan and audit the directory

Overview

Comprehensive audit system for AI coding assistant configurations: Content Type · Identification · Rule File Any Text/File · Pasted text or any file (any filename) · type-prompt.md AGENTS.md · Codex agent instructions · type-memory.md CLAUDE.md · Claude Code memory files · type-memory.md

Core Principles

Source: Based on Latest GPT Prompting Guide (openai-cookbook/examples/gpt-5)

Principle 0: GPT Prompting Guide Compliance (MANDATORY)

CRITICAL: This is the PRIMARY audit standard. Every audit MUST check these items and report findings. For ALL content containing AI instructions, verify: Check · What to Look For · Severity Verbosity constraints · Explicit length limits present · Severe Scope discipline · Explicit boundaries or prohibition list present · Severe

Fluxly profile Runtime and permissions are inferred by Fluxly static checks, not author declarations or runtime test results.
Domain
Security
Author-declared agents
No explicit declaration found; this is not inferred or tested compatibility
Static check
88 / 100 · heuristic scan, not runtime safety proof
Runtime evidence
Not run; no isolated runtime test evidence is currently recorded
Author / version / license
@tomevault-io · no license declared
Token usage
Heavy
Setup complexity
Guided setup
External API key
Not required
Operating systems
macOS · Linux · Windows
Runtime requirements
Node.js
Detected file/system behavior
  • Read-only
  • Write / modify
  • Shell exec
  • Env read
Network behavior
External requests
Install commands
None (reference only)

Heads up: Tool permissions are not declared; actual access depends on the runtime.

01
Step 0: Fetch Latest Prompting Guide (MANDATORY STANDARD)

CRITICAL: The GPT Prompting Guide is a primary audit standard, not just a reference. All prompts/instructions MUST be evaluated against these rules. Access this directory page to get file list: `https://github.com/openai/openai-cookbook/tree/main/examples/gpt-5` Select the latest version prompting guide (e.g., `gpt-5-2_prompting_guide.ipynb` > `gpt-5-1_prompting_guide.ipynb` > `gpt-5_prompting_guide.ipynb`)

02
Step 1: Detection & Classification

Scan path → identify type → load appropriate rules:

03
Step 2: Execute Universal Checks (ALL TYPES)

FIRST: Execute Principle 0 (GPT Guide Compliance) checks before proceeding. GPT Guide Compliance Check (MANDATORY FIRST): Execute each check from Principle 0 table, record status and evidence (line numbers, quotes). Every audit MUST execute these checks from `rules-universal.md`: Category · Action Required · Evidence Output

04
Step 3: Execute Type-Specific Checks

Based on content type, execute ALL checks in the relevant file: For Prompts (`type-prompt.md`): Check Category · Action Structure Validation · Verbosity constraints? Scope boundaries? Output format? Content Quality · Specific instructions? Not vague?

05
Step 4: Execute Cross-Cutting Checks (Multi-file Systems)

For Skills, Plugins, Composites, execute ALL checks from: From `cross-design-coherence.md`: Check · Action Full Directory Scan · Enumerate ALL files, classify each, build rule inventory Design Philosophy · Extract principles from all files, check consistency

06
Step 5: Issue Verification (5-Point Check)

For each suspected issue, verify ALL points: Concrete scenario - Can describe specific failure? Design scope - Within intended boundaries? Functional capability - Does implementation match claimed capability? Flaw vs choice - Unintentional error or valid design?

07
Step 6: Fix Proposal Verification (Principle Check)

CRITICAL: Before outputting ANY fix proposal, verify it against core principles: Check · Question · If NO → Occam's Razor · Is this addition truly necessary? Could the goal be achieved by DELETE/MERGE/MODIFY instead of ADD? · Reconsider fix approach AI Inference · Can AI infer the correct behavior from existing examples/context/patterns? · Do NOT add explicit rule Hardcoding Check · Is this adding hardcoded values (e.g., "≤5 bullets", "≤200 words") where AI should judge based on context? · Remove hardcoded values

08
Step 7: Generate Report

Follow `references/ref-output-format.md` for structure. Section 2 Cross-Cutting Analysis MUST include: Naming & Numbering: actual check results with specific findings TOC-Content Match: comparison results Reference Integrity: broken refs listed

09
Step 8: Wait for User Confirmation (PHASE GATE)

CRITICAL: After generating the report, STOP and wait for user input. Do NOT apply any fixes automatically. User interaction flow: Output complete audit report (Sections 0-5) STOP - Wait for user to select which fixes to apply Only after user confirms (e.g., "1", "1,2", "all") → Apply selected fixes

Discussion

Powered by GitHub Discussions. Sign in with GitHub to comment, react, or subscribe.