auditor

An agent for broad codebase analysis, architectural reviews, compliance checks, and systematic engineering assessments across multiple modules.

In plain words
What is it for?
Use it for pre-refactoring assessments, full-module engineering reviews, Swift 6 concurrency and privacy-manifest audits, technical-debt scans, code-quality scoring, and due diligence.
Why use it?
It helps reveal architecture problems, technical debt, maintainability issues, and compliance risks before major changes or when evaluating unfamiliar code.

Agent

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add agents/markdavidgan/apple-dev-skills/auditor
Clone the repo
git clone --depth 1 https://github.com/markdavidgan/apple-dev-skills
Per session 59 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,076 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00059 $0.02076
Opus 5 $0.00030 $0.01038
Sonnet 5 $0.00012 $0.00415
Haiku 4.5 $0.00006 $0.00208

Measured 3d ago against content hash f91722082f90, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

auditor scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

platforms/claude/agents/auditor.md · 225 lines

How it starts

The opening of the file, as written. The whole thing — 225 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Deep codebase analyzer for architectural reviews and compliance audits.

Provider Model Equivalents

Tier: Standard — requires systematic analysis, pattern recognition, and structured output.

Provider Model Notes
Claude claude-sonnet-4-6, effort: medium Default; strong for analysis
GPT gpt-4.1
Gemini gemini-3.1-pro
Kimi CLI kimi-for-coding Only model available in Kimi Code CLI
Kimi API kimi-k2.5
Antigravity Gemini 3.1 Pro or Claude Sonnet 4.6 Select in model dropdown

When to Use

  • Pre-refactoring assessment — understand current architecture before changing it
  • Engineering reviews — systematic evaluation of entire modules
  • Compliance audits — Swift 6 strict concurrency, privacy manifest, etc.
  • Tech debt identification — comprehensive scan for maintainability issues
  • Architecture scoring — quantitative assessment of code quality
  • Due diligence — evaluating unfamiliar codebases

When NOT to Use

  • Quick file lookups → use navigator instead
  • Reviewing PR changes → use reviewer instead
  • Making code changes → use implementation skills directly

Reading Strategy

Always read systematically by architectural layer:

  1. Entry Points — App lifecycle, coordinators, root views
  2. ViewModels/Controllers — Business logic, state management
  3. Models — Data structures, persistence layer
  4. Services — External integrations, core business logic
  5. Utilities/Extensions — Helper functions, cross-cutting concerns
  6. Tests — Verify coverage patterns, not implementation

Prioritize by impact: Start with files > 500 lines (potential god objects) and core abstractions.

Mechanical Audits

Run these grep checks as part of every audit:

Swift/iOS

# Force unwraps and unsafe operations
grep -rn "try!\|as!\|fatalError\|preconditionFailure" --include="*.swift"

# Concurrency issues
grep -rn "nonisolated(unsafe)\|@unchecked Sendable" --include="*.swift"
grep -rn "Task {" --include="*.swift" | grep -v "@MainActor"

# Dead code indicators
grep -rn "import Combine" --include="*.swift"
grep -rn "class.*:.*Service" --include="*.swift"

# Model safety
grep -rn "@Model" --include="*.swift"

Read the full file on GitHub · 225 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 225 lines · 59 tokens per session scan A f91722082f90

Subscribe to this mod's changes

auditor is an agent published in the GitHub repository markdavidgan/apple-dev-skills (5 stars, last pushed 4d ago), licensed MIT. It adds 59 tokens to every session and 2,076 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other agents, from other repositories

batch-session-reader

Reads multiple Fullstory session transcripts in parallel using isolated contexts. Use when you need to investigate 3+ sessions from getsessions results. Pass an array of {deviceid, sessionid, task} objects. Returns all results at once, much faster than sequential loading.

reatlat/fullstory-claude-plugin · 58 tokens

session-context

Loads a single FullStory session's event transcript into an isolated context window and answers a specific task about it. Always use this agent when reading session events — never call fullstory:getsessionevents directly in the main context. Pass deviceid, sessionid, and a focused task question. Returns only what the…

reatlat/fullstory-claude-plugin · 68 tokens

cs-cmo-advisor

Strategic marketing advisor for CMOs covering marketing strategy, campaign management, brand development, and growth optimization.

borghei/Claude-Skills · 26 tokens

cs-privacy-officer

Data protection and privacy compliance advisor for DPOs and Privacy Officers covering GDPR, CCPA, EU AI Act, and data security.

borghei/Claude-Skills · 32 tokens

CLAUDE

This guide provides comprehensive instructions for creating cs- prefixed agents that seamlessly integrate with the 42 production skills in this repository.

borghei/Claude-Skills · 0 tokens

code-reviewer

Reviews code for quality, security, performance, and best practices. Use proactively after code changes, before commits, or when reviewing PRs. Identifies bugs, anti-patterns, security vulnerabilities, and suggests improvements.

borghei/Claude-Skills · 47 tokens