NWave nw-review
Dispatches an expert reviewer agent to critique workflow artifacts. Use when a roadmap, implementation, or step needs quality review before proceeding.
git clone https://github.com/nWave-ai/nWave
T=$(mktemp -d) && git clone --depth=1 https://github.com/nWave-ai/nWave "$T" && mkdir -p ~/.claude/skills && cp -r "$T/nWave/skills/nw-review" ~/.claude/skills/nwave-ai-nwave-nw-review && rm -rf "$T"
nWave/skills/nw-review/SKILL.mdNW-REVIEW: Expert Critique and Quality Assurance
Wave: CROSS_WAVE Agent: Dynamic (nw-*-reviewer)
Overview
Dispatches expert reviewer agent to critique workflow artifacts. Takes base agent name, appends
-reviewer, invokes with artifact. Reviewer agent owns all review methodology|criteria|output format.
Review Philosophy: Radical Candor
Every review MUST embody Radical Candor — kind AND clear, specific AND sincere:
- Care personally: Acknowledge what works. Understand author's intent before critiquing. Include at least one genuine
per review.praise: - Challenge directly: Be specific about what is wrong and WHY. Ground feedback in evidence|consequences, not preference. Never soften security/data-loss issues.
- Avoid ruinous empathy: Never "LGTM" when real issues exist. Hedging ("maybe consider possibly...") on blocking concerns is a review failure.
- Avoid obnoxious aggression: Never "this is terrible" without constructive alternative. Focus on work, not author. Explain "why" behind every critique.
Feedback Format: Conventional Comments
All findings MUST use Conventional Comments labels:
| Label | Purpose | Blocking? |
|---|---|---|
| Highlight something done well (genuine, not filler) | No |
| Must be resolved before proceeding | Yes |
| Security vulnerability — maximum directness | Yes |
| Propose improvement with reasoning | Mark or |
| Trivial, preference-based | No |
| Seek clarification before assuming | No |
| Idea sparked by the review | No |
Findings MUST be priority-ordered: blocking issues first, then suggestions, then nitpicks/praise.
Approval Criteria
| Verdict | Criteria |
|---|---|
| APPROVED | No blocking issues. Non-blocking feedback is advisory. |
| NEEDS_REVISION | Blocking issues exist. Author must address. Each blocking issue enumerated. |
| REJECTED | Fundamental design problems requiring significant rework. Rare — explain thoroughly, offer alternatives. |
Syntax
/nw-review @{agent-name} {artifact-type} "{artifact-path}" [step_id={id}] [--dimensions=rpp] [--from=1] [--to=3]
Parameters:
- Base agent (e.g.,@{agent-name}
).@nw-software-crafter
suffix appended automatically.-reviewer
- One of:{artifact-type}
,baseline
,roadmap
,step
,taskimplementation
- Path to artifact file (resolved to absolute){artifact-path}
- Required for step and implementation reviewsstep_id={id}
- Triggers RPP code smell scan alongside standard review (Dimension 4)--dimensions=rpp
/--from=N
- RPP level range (default: 1-6). Requires--to=N--dimensions=rpp
Rigor Profile Integration
Before dispatching the reviewer agent, read rigor config from
.nwave/des-config.json (key: rigor). If absent, use standard defaults.
: Ifreview_enabled
, skip the review entirely. Output: "Review skipped per rigor profile (review_enabled=false)."false
: Pass asreviewer_model
parameter to Task tool. Ifmodel
, skip the review. Overrides the default Haiku model."skip"
: Ifdouble_review
and called from deliver Phase 4, the caller is responsible for invoking review twice.true
Agent Derivation
| User provides | Reviewer invoked |
|---|---|
| |
| |
| |
Default model: Haiku (overridden by
rigor.reviewer_model when set).
Agent Invocation
- Parse parameters — Strip
from agent name, resolve artifact path to absolute, extract optional step_id, dimensions, from/to range. Gate: all parameters parsed.@ - Read rigor config — Read
key.nwave/des-config.json
. If absent, use standard defaults. Gate: rigor profile loaded or defaults applied.rigor - Validate inputs — Run all four validation checks below. Gate: zero validation failures.
- Apply rigor overrides — Check
(skip if false), determine model fromreview_enabled
(default: haiku, skip if "skip"). Gate: execution decision made.reviewer_model - Invoke reviewer — Call Task tool with
, resolved model, and promptsubagent_type="{agent-name}-reviewer"
. Reviewer handles reading artifact, applying domain expertise, generating structured critique, updating original artifact with review metadata. Gate: Task tool invoked."Review {artifact-type}: {absolute-artifact-path} [step_id={id}]"
Validation (before invoking)
- Agent exists — Strip
, check agent name against agent registry. Gate: agent found or return "Unknown agent: {name}. Check available agents with /nw-agents."@ - Artifact type valid — Confirm type is one of: baseline, roadmap, step, task, implementation. Gate: type valid or return "Invalid artifact type: {type}. Use: baseline, roadmap, step, task, implementation."
- Artifact file exists — Resolve to absolute path and confirm file exists. Gate: file found or return "Artifact not found: {path}."
- step_id present when required — Require step_id when artifact type is
orstep
. Gate: step_id provided or return "step_id required for {type} reviews."implementation
Success Criteria
- Reviewer agent invoked (not self-performed)
- Original artifact file updated with review metadata
- Review includes severity levels and approval status (APPROVED, NEEDS_REVISION, REJECTED)
Examples
Example 1: Step review
/nw-review @nw-software-crafter step "docs/feature/auth-upgrade/execution-log.json" step_id=02-01
Invokes
nw-software-crafter-reviewer with step review of execution log, step 02-01.
Example 2: Roadmap review
/nw-review @nw-solution-architect roadmap "docs/feature/auth-upgrade/roadmap.json"
Invokes
nw-solution-architect-reviewer with roadmap review.
Example 3: Implementation review
/nw-review @nw-platform-architect implementation "docs/feature/auth-upgrade/execution-log.json" step_id=01-01
Invokes
nw-platform-architect-reviewer with implementation review of step 01-01.
Example 4: RPP code quality review
/nw-review @nw-software-crafter implementation "src/des/" --dimensions=rpp --from=1 --to=3
Invokes
nw-software-crafter-reviewer with implementation review + RPP L1-L3 code smell detection using cascade rule.
Error Messages
- Invalid agent: "Unknown agent: {name}. Check available agents with /nw-agents."
- Invalid type: "Invalid artifact type: {type}. Use: baseline, roadmap, step, task, implementation."
- Missing file: "Artifact not found: {path}."
- Missing step_id: "step_id required for {type} reviews."
Next Wave
Handoff To: Depends on review outcome (rework or proceed) Deliverables: Updated artifact file with embedded review metadata
Expected Outputs
Updated artifact file (roadmap.json, execution-log.json, etc.) with reviews section