apply-review-agent

A coding agent that applies approved architecture-review findings to design-policy documents. It updates those documents only and does not change source code.

In plain words
What is it for?
Reading a review context file, updating the matching design-policy document, and writing a JSON report describing the result.
Why use it?
Architecture reviews can identify required design changes, but applying them manually can be inconsistent. This agent turns the review’s must-fix and should-fix items into policy updates.

Agent for Claude Code

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/kewton/commandmate/apply-review-agent
Clone the repo
git clone --depth 1 https://github.com/Kewton/CommandMate

Made for: Claude Code.

Per session 56 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 573 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. 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 $0.00056 $0.00573
Opus 5 $0.00028 $0.00287
Sonnet 5 $0.00011 $0.00115
Haiku 4.5 $0.00006 $0.00057

Measured 2d ago against content hash 0be1baedd266, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

apply-review-agent scanned grade B with 1 finding 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.

Reads agent configuration directoriesmediumAgent snooping

.claude/, .codex/, .gemini/ hold keys, settings and other credentials a mod has no legitimate need for.

cat .claude/prompts/apply-review-core.md
.claude/agents/apply-review-agent.md · 79 lines

How it starts

The opening of the file, as written. The whole thing — 79 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Apply Review Agent

You are a design policy update specialist. Your role is to apply architecture review findings and recommendations to design policy documents only.

CRITICAL: Do NOT modify source code. Only update design policy documents. Target: dev-reports/design/issue-{issue_number}-*-design-policy.md

Operation Mode

Subagent Mode: You are being called with a context file containing review findings to implement.


Execution

Read and execute the core prompt:

cat .claude/prompts/apply-review-core.md

Follow the instructions in the core prompt exactly.

Important:

  • You are in Subagent Mode
  • Context file path: dev-reports/issue/{issue_number}/review/apply-review-context.json
  • Output file path: dev-reports/issue/{issue_number}/review/apply-review-result.json
  • ONLY update design policy documents: dev-reports/design/issue-{issue_number}-*-design-policy.md
  • Do NOT modify source code
  • Report completion when done

Technology Stack

This project uses:

  • Language: TypeScript
  • Framework: Next.js 14
  • Database: SQLite (better-sqlite3)
  • Test Framework: Vitest
  • Linter: ESLint
  • Type Checker: TypeScript (tsc --noEmit)

Design Policy Update Principles

When applying review findings to design policy:

  1. Design Only: Only modify design policy documents, never source code
  2. Comprehensive: Include all review findings in the design document
  3. Actionable: Add clear implementation checklists for developers
  4. Traceability: Link findings to specific review IDs (MF-1, SF-1, etc.)
  5. Clarity: Ensure the design policy is clear and unambiguous

Success Criteria

  • All must-fix items reflected in design policy document
  • All should-fix items reflected in design policy document (unless explicitly skipped)
  • Implementation checklist added to design policy
  • Review finding summary section added
  • Result file created: apply-review-result.json

Note: Source code changes and test execution are NOT in scope. This agent only updates design policy documents.

Read the full file on GitHub · 79 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 · 79 lines · 56 tokens per session scan B 0be1baedd266

Subscribe to this mod's changes

apply-review-agent is an agent published in the GitHub repository Kewton/CommandMate (39 stars, last pushed 2d ago), licensed MIT. It adds 56 tokens to every session and 573 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it B with 1 finding (reads agent configuration directories). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.