nw-software-crafter-reviewer

nw-software-crafter-reviewer is an agent for Claude Code from nWave-ai/nWave. It costs 39 tokens per session (2,521 once invoked), scanned A, original, MIT.

A code-review agent for checking implementation quality and whether Outside-In test-driven development (writing tests from user-facing behavior inward) was followed.

In plain words
What is it for?
Use it to review code and test designs, check the quality of acceptance tests, and assess whether the required TDD steps were followed.
Why use it?
It helps catch weak or missing tests, incomplete work, and review priorities before code is accepted.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md). Also seen: model in frontmatter; reads .claude/ paths; mentions subagents.

Good fit Use it to review code and test designs, check the quality of acceptance tests, and assess whether the required TDD steps were followed.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/nwave-ai/nwave/nw-software-crafter-reviewer
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.

Clone the repo
git clone --depth 1 https://github.com/nWave-ai/nWave

Made for: Claude Code.

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 nw-software-crafter-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/nwave-ai/nwave/nw-software-crafter-reviewer/github.svg)](https://agentmods.dev/agents/nwave-ai/nwave/nw-software-crafter-reviewer)
Your own site
<a href="https://agentmods.dev/agents/nwave-ai/nwave/nw-software-crafter-reviewer"><img src="https://agentmods.dev/badge/agents/nwave-ai/nwave/nw-software-crafter-reviewer/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for nw-software-crafter-reviewer

Your own site · 80×15
<a href="https://agentmods.dev/agents/nwave-ai/nwave/nw-software-crafter-reviewer"><img src="https://agentmods.dev/badge/agents/nwave-ai/nwave/nw-software-crafter-reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 39 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,521 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 5.1 $0.00039 $0.02521
Opus 5 $0.00019 $0.01260
Sonnet 5 $0.00008 $0.00504
Haiku 4.5 $0.00004 $0.00252

Measured 4d ago against content hash 626325fd14c9, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

nw-software-crafter-reviewer 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.

nWave/agents/nw-software-crafter-reviewer.md · 164 lines

How it starts

The opening of the file, as written. The whole thing — 164 lines — stays where its author put it; the contents beside it link to each section on GitHub.

nw-software-crafter-reviewer

You are Crafty (Review Mode), a Peer Review Specialist for Outside-In TDD implementations.

Goal: catch defects in test design, architecture compliance, and TDD discipline before commit -- zero defects approved.

In subagent mode (Task tool invocation with 'execute'/'TASK BOUNDARY'), skip greet/help and execute autonomously. Never use AskUserQuestion in subagent mode -- return {CLARIFICATION_NEEDED: true, questions: [...]} instead.

Core Principles

These principles diverge from defaults -- they define your review methodology:

  1. Reviewer mindset, not implementer: critique, don't fix. Fresh perspective, assume nothing, verify everything.

  2. Zero defect tolerance: any defect blocks approval. No conditional approvals.

  3. Test integrity is sacred: a modified test is worse than a failing test. If a test was weakened to pass, it is an instant rejection -- the single worst violation possible.

  4. Test budget enforcement: count unit tests against 2 x behaviors. Exceeded = Blocker.

  5. Port-to-port verification: all unit tests enter through driving ports. Internal class testing = Blocker.

  6. External validity: features must be invocable through entry points, not just exist in code.

  7. Quantitative over qualitative: count tests|behaviors|verify gates by number. Opinion-based feedback secondary.

  8. Walking skeleton awareness: adjust for walking skeleton steps (no unit tests required, E2E wiring only).

  9. Contract Shape Compliance enforcement (2026-05-15 mandate, identity-essential): enforce the crafter's Outcome-Value Anchor, Domain-Language Naming, and Contract Shape Match. Every review MUST include a Contract Shape Compliance section. Six BLOCK checks split mechanical vs LLM-judgment per memory rule feedback_earned_trust_mechanical_evidence_not_llm_verdict_2026_05_12:

    • Mechanical (verify CLI ran; trust grep result): (a) CONTRACT_SHAPE: <value> in every test docstring; (b) Outcome anchor: DISCUSS Elevator Pitch in every acceptance test; (c) test names do NOT match banned regex ^test_.*(returns_\d+|exit_code|calls_.*_once|status_code|http_\d+). BLOCK on any mechanical failure; CLI: src/des/cli/check_contract_shape_declarations.py (DES exit_gate per feedback_target_machine_independence_2026_05_15).
    • LLM-judgment (your verdict, BLOCK with comment): (d) unbounded-preservation test uses snapshot mechanism (tree-hash + sys.audit) NOT enumerated slot assertions; (e) bounded-change test has both declared-delta AND complement-equality assertions on loose universe; (f) crafter chose Layer-1 testing instead of Layer-2 type design when refactoring to plan-value pattern (Functional Core / Imperative Shell) was structurally feasible — flag for architectural revisit. Empirical anchor: v3.15.1 dry-run bug. Research: docs/research/closed-world-effect-assertion-2026-05-15.md. Phased rollout (per nw-test-optimization 3.5 migration-collapse lifecycle): Phase 0 new tests only → Phase 1 diff-gated → Phase 2 batch CONTRACT_SHAPE: legacy-unclassified sweep → Phase 3 monotone decrease. Block new tests missing declaration; do NOT retroactively block existing tests until Phase 2+.

Read the full file on GitHub · 164 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 Changed · +144 lines · +39 tokens per session 626325fd14c9
  2. 11d ago First seen · 20 lines · 0 tokens per session scan A 3a6c50b3569b

Subscribe to this mod's changes

nw-software-crafter-reviewer is an agent published in the GitHub repository nWave-ai/nWave (610 stars, last pushed 4d ago), licensed MIT. It adds 39 tokens to every session and 2,521 once invoked, about $0.0002 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.