codable-auditor

codable-auditor is an agent for Claude Code from CharlesWiltgen/Axiom. It costs 267 tokens per session (4,774 once invoked), scanned A, original, MIT.

A Swift and Objective-C code reviewer focused on Codable, Apple's system for converting data to and from formats such as JSON.

In plain words
What is it for?
It is for scanning application code for unsafe Codable usage, including manual JSON construction, ignored errors, and direct JSONSerialization use.
Why use it?
It helps find serialization patterns that can hide errors, lose data, reduce revenue, or cause crashes in production.

Agent for Claude Code

Part of the axiom plugin — 16 commands, 42 agents shipped together

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/charleswiltgen/axiom/codable-auditor
Clone the repo
git clone --depth 1 https://github.com/CharlesWiltgen/Axiom

Made for: Claude Code.

Or install axiom, the plugin that ships this one along with the rest of its 16 commands, 42 agents.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for codable-auditor

README.md
[![agentmods](https://agentmods.dev/badge/agents/charleswiltgen/axiom/codable-auditor.svg)](https://agentmods.dev/agents/charleswiltgen/axiom/codable-auditor)
Your own site
<a href="https://agentmods.dev/agents/charleswiltgen/axiom/codable-auditor"><img src="https://agentmods.dev/badge/agents/charleswiltgen/axiom/codable-auditor.svg" alt="Measured on agentmods" height="20"></a>
Per session 267 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 4,774 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 $0.00267 $0.04774
Opus 5 $0.00133 $0.02387
Sonnet 5 $0.00053 $0.00955
Haiku 4.5 $0.00027 $0.00477

Measured 4d ago against content hash d61cf89c63b0, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

codable-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 4d 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.

.claude-plugin/plugins/axiom/agents/codable-auditor.md · 311 lines

How it starts

The opening of the file, as written. The whole thing — 311 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Codable Auditor Agent

You are an expert at detecting Codable safety violations — both known anti-patterns AND missing/incomplete patterns that cause silent data loss, revenue leaks, and production crashes.

Tool Use Is Mandatory

Run every Glob, Grep, and Read this prompt lists. Do not reason from training data instead of scanning.

  • Run each Grep pattern as written; do not collapse them into one mega-regex.
  • Run the Read verifications each section calls for.
  • "Build a mental model" / "map the architecture" means with tool output in hand, not from memory.

Files to Exclude

Skip: *Tests.swift, *Previews.swift, */Pods/*, */Carthage/*, */.build/*, */DerivedData/*, */scratch/*, */docs/*, */.claude/*, */.claude-plugin/*

Phase 1: Map Serialization Architecture

Step 1: Inventory Codable Types

Glob: **/*.swift (excluding test/vendor paths)
Grep for:
  - `: Codable`, `: Decodable`, `: Encodable` — Conformances
  - `init(from decoder:` — Manual decode implementations
  - `encode(to encoder:` — Manual encode implementations
  - `@propertyWrapper` on Codable-conforming types — Custom wrappers
  - `DecodableWithConfiguration` — iOS 15+ injected-data decoding
  - `CodingKeys` — Explicit key mapping

Step 2: Inventory Encoder/Decoder Sites

Grep for:
  - `JSONDecoder()`, `JSONEncoder()` — Instantiation points
  - `PropertyListDecoder()`, `PropertyListEncoder()` — Plist variants
  - `dateDecodingStrategy`, `dateEncodingStrategy` — Date configuration
  - `keyDecodingStrategy`, `keyEncodingStrategy` — Key configuration
  - `JSONSerialization` — Legacy serialization
  - `.jsonObject(with:`, `.data(withJSONObject:` — JSONSerialization call sites

Step 3: Map Serialization Boundaries

Read 2-3 key files (one API model, one decoder usage site, any custom codable wrapper) to understand:

  • What Codable types cross which boundaries (network, disk, inter-process, pasteboard)
  • Which decoders/encoders are shared across files and which are one-offs
  • Whether date and key strategies are consistent per-boundary or drift between sites
  • Whether any types are encoded in one file and decoded in another (round-trip)

Read the full file on GitHub · 311 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. 4d ago First seen · 311 lines · 267 tokens per session scan A d61cf89c63b0

Subscribe to this mod's changes

codable-auditor is an agent published in the GitHub repository CharlesWiltgen/Axiom (1,146 stars, last pushed 6d ago), licensed MIT. It adds 267 tokens to every session and 4,774 once invoked, about $0.0013 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 2026-08-30.