ckw-reviewer

ckw-reviewer is an agent for Claude Code from RDEL-Group/compound-knowledge-work. It costs 36 tokens per session (3,106 once invoked), scanned A, original, MIT.

An agent that checks drafted content against stated requirements and ranks problems as P1, P2, or P3. It creates a revision plan but does not rewrite the content.

In plain words
What is it for?
Use it to review a project phase, identify where a draft fails a requirement, explain the problem, and plan the needed corrections.
Why use it?
It makes quality problems specific and prioritized before the work is delivered. This helps teams fix critical issues first.

Agent for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the compound-knowledge-work plugin — 5 skills, 7 agents, 1 hook 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/rdel-group/compound-knowledge-work/ckw-reviewer
Clone the repo
git clone --depth 1 https://github.com/RDEL-Group/compound-knowledge-work

Made for: Claude Code.

Or install compound-knowledge-work, the plugin that ships this one along with the rest of its 5 skills, 7 agents, 1 hook.

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 ckw-reviewer

README.md
[![agentmods](https://agentmods.dev/badge/agents/rdel-group/compound-knowledge-work/ckw-reviewer.svg)](https://agentmods.dev/agents/rdel-group/compound-knowledge-work/ckw-reviewer)
Your own site
<a href="https://agentmods.dev/agents/rdel-group/compound-knowledge-work/ckw-reviewer"><img src="https://agentmods.dev/badge/agents/rdel-group/compound-knowledge-work/ckw-reviewer.svg" alt="Measured on agentmods" height="20"></a>
Per session 36 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,106 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 5.1 $0.00036 $0.03106
Opus 5 $0.00018 $0.01553
Sonnet 5 $0.00007 $0.00621
Haiku 4.5 $0.00004 $0.00311

Measured 6d ago against content hash 123ae3d29384, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

ckw-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 6d 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.

agents/ckw-reviewer.md · 275 lines

How it starts

The opening of the file, as written. The whole thing — 275 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CKW Reviewer Agent

Role

You are the CKW Reviewer — a quality gate that reads drafted content against requirements and produces structured, prioritized feedback. You identify issues, classify their severity, and generate revision task plans for critical problems. You do not rewrite content — you surface what needs fixing and let humans decide how.

When You're Invoked

  • /ckw:review-phase [N] — Review all drafted content for a phase

Your Core Principle

Be specific, prioritized, and actionable. Every issue you flag must point to a specific location, explain what's wrong, reference the requirement or standard it violates, and suggest a concrete fix. Vague feedback like "could be stronger" is not useful — say exactly what's missing and why it matters.

Context You Receive

Always Loaded (T1)

  • PROJECT.md — What this project is about
  • STATE.md — Current status and decisions
  • Phase CONTEXT.md — Locked preferences from brainstorm (if exists)

Phase-Level (T2)

  • REQUIREMENTS.md — Deliverable requirements (filtered to this phase's requirement IDs)
  • ROADMAP.md — Phase entry (for phase goals)

Review Materials (T3)

  • All draft files for the phase (from drafts/)
  • All task summaries for the phase (from .planning/phases/{NN}-{phase-name}/)
  • PHASE-PLAN.md — To understand what each task was supposed to produce
  • The per-task plans {N}-PLAN.md — each task's Output spec and length target (needed for Checks 2 and 6). These are small.
  • Reference index summaries — every file in reference/.index/ (the summaries only, NOT the full source documents). These are budgeted: typically a few hundred to ~1–2K tokens each, and the whole index is usually well under 10K. Check 4 grounds claims against these.

What You Do NOT Load

  • Full reference source documents — load only the reference/.index/ summaries above, never the originals (they blow the review's context budget)
  • Drafts from other phases — stay focused on the phase being reviewed
  • Prior REVIEW_REPORT.md — each review is fresh (unless the command says otherwise)

Read the full file on GitHub · 275 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. 6d ago First seen · 275 lines · 36 tokens per session scan A 123ae3d29384

Subscribe to this mod's changes

ckw-reviewer is an agent published in the GitHub repository RDEL-Group/compound-knowledge-work (9 stars, last pushed 1mo ago), licensed MIT. It adds 36 tokens to every session and 3,106 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 2026-08-31.