code-review

A method for reviewing code written by an AI coding agent, with attention to architecture, design, engineering practices, and test quality. It starts by examining the project structure before tracing how data and behavior move through the code.

In plain words
What is it for?
Use it to assess domain models, module boundaries, encapsulation, business-logic placement, and whether tests verify real behavior.
Why use it?
It provides an evidence-based way to judge code instead of assuming the author’s intentions or rewarding code merely because tests exist. This helps reveal weak boundaries, exposed internals, and shallow tests.

Skill for Claude CodeCodex

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 skills/noesisvision/nasde-toolkit/code-review
Any agent
npx skills add NoesisVision/nasde-toolkit --skill code-review
Clone the repo
git clone --depth 1 https://github.com/NoesisVision/nasde-toolkit

Made for: Claude Code, Codex.

Per session 19 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 495 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 $0.00019 $0.00495
Opus 5 $0.00010 $0.00247
Sonnet 5 $0.00004 $0.00099
Haiku 4.5 $0.00002 $0.00049

Measured 2d ago against content hash 1df3b8ad29b5, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

code-review 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.

examples/ddd-architectural-challenges/evaluator_skills/code-review/SKILL.md · 45 lines

How it starts

The opening of the file, as written. The whole thing — 45 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Code Review for Assessment Evaluation

You are reviewing code produced by an AI coding agent. Your goal is to provide precise, evidence-based scoring — not to be lenient or harsh, but accurate.

Review methodology

  1. Start with structure — Glob to understand the file tree before reading individual files. The shape of the codebase tells you about architectural decisions.

  2. Read critically, not charitably — Score what IS there, not what the author probably meant. If a pattern is half-implemented, score it as half-implemented.

  3. Trace the domain model — Follow the flow from entry point to persistence. Look for:

    • Are domain concepts explicit types or buried in primitives?
    • Do boundaries between modules/layers exist and hold?
    • Is business logic in the domain or scattered across infrastructure?
  4. Check encapsulation — Look for:

    • Public fields that should be private
    • Getter/setter pairs that expose internals
    • Domain objects that are just data bags with no behavior
    • Invariants that are enforced externally rather than internally
  5. Evaluate test quality — Tests that merely exist are not enough. Check:

    • Do tests verify behavior or just call methods?
    • Are edge cases and failure modes covered?
    • Do test names describe the scenario being tested?
    • Are tests testing the unit or the framework?
  6. Look for anti-patterns — Common problems to flag:

    • Anemic domain models (logic in services, entities are just DTOs)
    • Leaking abstractions (domain depends on infrastructure types)
    • God classes or methods doing too many things
    • Copy-paste with minor variations instead of proper abstraction

Scoring principles

  • Evidence required — Every score must cite specific files, classes, or code patterns. "The code generally looks good" is not evidence.
  • Calibration — A max score means excellent, not merely acceptable. Reserve top scores for genuinely well-crafted code.
  • Partial credit — If the agent solved the core problem but cut corners on secondary concerns, reflect both in the score and reasoning.
  • Zero scores are valid — If a dimension was completely ignored by the agent, score 0 with explanation.

Read the full file on GitHub · 45 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 · 45 lines · 19 tokens per session scan A 1df3b8ad29b5

Subscribe to this mod's changes

code-review is a skill published in the GitHub repository NoesisVision/nasde-toolkit (12 stars, last pushed 8d ago), licensed MIT. It adds 19 tokens to every session and 495 once invoked, about $0.0001 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.