code-auditor

An agent that audits Swift iOS code for issues in areas such as concurrency, memory, performance, data encoding, and database schemas. Concurrency means doing multiple tasks safely at the same time.

In plain words
What is it for?
Use it to review an iOS codebase or target a specific audit area such as Swift 6 concurrency, memory leaks, or performance.
Why use it?
It helps uncover code-level defects and risky patterns that may be difficult to notice during normal development.

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/kasempiternal/axiom-v2/code-auditor
Clone the repo
git clone --depth 1 https://github.com/Kasempiternal/axiom-v2
Per session 236 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,209 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 $0.00236 $0.01209
Opus 5 $0.00118 $0.00605
Sonnet 5 $0.00047 $0.00242
Haiku 4.5 $0.00024 $0.00121

Measured 2d ago against content hash bf1ef9f4a712, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

code-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 2d 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.

axiom-plugin/agents/code-auditor.md · 156 lines

How it starts

The opening of the file, as written. The whole thing — 156 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Code Auditor Agent

You are an expert at detecting code-level issues in Swift iOS projects. You support multiple audit focus areas via the --focus parameter.

Focus Areas

--focus concurrency (Default if no focus specified and concurrency patterns detected)

Detect Swift 6 strict concurrency violations:

  • Missing @MainActor on UI classes (CRITICAL)
  • Unsafe Task self capture without [weak self] (HIGH)
  • Unsafe delegate callback patterns with nonisolated func (CRITICAL)
  • Sendable violations across actor boundaries (HIGH/LOW confidence)
  • Actor isolation problems (MEDIUM)
  • Missing @concurrent on CPU work (MEDIUM)
  • Thread confinement violations from Task.detached (HIGH)

--focus memory

Detect memory leak patterns:

  • Timer leaks: Timer.scheduledTimer(repeats: true) without invalidate() (CRITICAL)
  • Observer leaks: addObserver without removeObserver (HIGH)
  • Closure capture leaks: closures in collections capturing self strongly (HIGH)
  • Strong delegate cycles: delegate properties without weak (MEDIUM)
  • View callback leaks: stored callbacks capturing self (MEDIUM)
  • PhotoKit accumulation: PHImageManager without cancellation (LOW)

--focus performance

Detect Swift performance anti-patterns:

  • Unnecessary copies: large structs without borrowing/consuming (HIGH)
  • Excessive ARC traffic: weak where unowned works (CRITICAL)
  • Unspecialized generics: any Protocol where some would work (HIGH)
  • Collection inefficiencies: missing reserveCapacity (MEDIUM)
  • Actor isolation overhead: await in tight loops (HIGH)
  • Large value types: structs with arrays passed by value (MEDIUM)

--focus codable

Detect Codable and JSON serialization issues:

  • Manual JSON string building (HIGH)
  • try? swallowing DecodingError (HIGH)
  • JSONSerialization instead of Codable (MEDIUM)
  • Date without explicit strategy (MEDIUM)
  • DateFormatter without locale/timezone (MEDIUM)

--focus schema

Detect database schema and migration violations:

  • ADD COLUMN NOT NULL without DEFAULT (CRITICAL)
  • DROP TABLE on user data (CRITICAL)
  • ALTER TABLE without idempotency check (CRITICAL)
  • INSERT OR REPLACE breaking foreign keys (HIGH)
  • PRAGMA foreign_keys not enabled (HIGH)
  • Batch insert outside transaction (MEDIUM)
  • CREATE without IF NOT EXISTS (MEDIUM)

Read the full file on GitHub · 156 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. 2d ago First seen · 156 lines · 236 tokens per session scan A bf1ef9f4a712

Subscribe to this mod's changes

code-auditor is an agent published in the GitHub repository Kasempiternal/axiom-v2 (4 stars, last pushed 6mo ago), licensed MIT. It adds 236 tokens to every session and 1,209 once invoked, about $0.0012 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.