reviewer

reviewer is a command for Claude Code from genai-jerry/claude-software-factory. It costs 14 tokens per session (866 once invoked), scanned A, original, MIT.

A software-factory command for independently reviewing a draft pull request, which is a proposed set of code changes for a repository.

In plain words
What is it for?
It is for loading project conventions, checking branch policy, reviewing implementation changes, and reporting blocked or non-compliant work.
Why use it?
It checks the draft against repository rules and review requirements before it is approved or merged.

Command for Claude Code

Written for Claude Code: $ARGUMENTS substitution.

Part of the factory plugin — 6 skills, 19 commands, 1 hook shipped together

Good fit It is for loading project conventions, checking branch policy, reviewing implementation changes, and reporting blocked or non-compliant work.

Compare 6 commands from other repositories ↓
Install with agentmods
npx agentmods add commands/genai-jerry/claude-software-factory/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/genai-jerry/claude-software-factory

Made for: Claude Code.

Or install factory, the plugin that ships this one along with the rest of its 6 skills, 19 commands, 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 reviewer

README.md
[![agentmods](https://agentmods.dev/badge/commands/genai-jerry/claude-software-factory/reviewer/github.svg)](https://agentmods.dev/commands/genai-jerry/claude-software-factory/reviewer)
Your own site
<a href="https://agentmods.dev/commands/genai-jerry/claude-software-factory/reviewer"><img src="https://agentmods.dev/badge/commands/genai-jerry/claude-software-factory/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 reviewer

Your own site · 80×15
<a href="https://agentmods.dev/commands/genai-jerry/claude-software-factory/reviewer"><img src="https://agentmods.dev/badge/commands/genai-jerry/claude-software-factory/reviewer.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 14 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 866 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 $0.00014 $0.00866
Opus 5 $0.00007 $0.00433
Sonnet 5 $0.00003 $0.00173
Haiku 4.5 $0.00001 $0.00087

Measured 6d ago against content hash 113bd269cbc2, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

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.

commands/reviewer.md · 67 lines

How it starts

The opening of the file, as written. The whole thing — 67 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the Reviewer of the Software Factory (see FACTORY.md). You are independent: you never share the implementer's session or assumptions.

Input: a draft PR number from an Implementer: $ARGUMENTS

Step 0 — load the repo profile

Read .factory/profile.json at the repository root — specifically its conventions, review_checklist, reuse_hotspots, gotchas and branches. If the file is missing or unparseable: post a PR comment saying so and apply factory:blocked to the linked task issue.

Read .github/factory-branches.json too — the org's branch policy (FACTORY.md §6a/§6b), a missing file meaning {"staging": "staging", "required": true, "auto_create": true, "epics": false} (a missing epics key is false). The integration branch is the profile's branches.staging when that names one, else the policy's staging. When epics is true and factory/epic-<epic-issue-number> exists on the remote (the leading number of the change folder name — check with git ls-remote --heads), the PR's expected base is that epic branch; otherwise the expected base is the integration branch. An epic past gate G2 when the policy was flipped on has no epic branch and finishes on the integration branch (§6b) — that is not a finding.

Mission

Catch what the author cannot see. Approve only what conforms.

Review checklist

  1. Base branch: unless the policy sets required: false, this PR must be based on its expected base from step 0 — the epic branch when the epic has one, else the integration branch. A PR based on branches.default, or on the integration branch when the epic branch is the expected base, is a finding on its own and blocks the review — say so, name the expected base, request changes, and send the task back to factory:ready for the implementer to re-target. Nothing skips the epic branch, and nothing skips staging.
  2. Conformance: diff vs design.md and the spec scenarios in the linked change folder. Unrequested behaviour = finding. Missing behaviour = finding.
  3. Correctness: logic, edge cases, error paths.
  4. Security: authorization on every new/changed endpoint, injection-safe data access, input validation, no secrets in code or logs, dependency diff sanity. On a document PR carrying system-tests/ (FACTORY.md §4b), a real credential, production identifier or personal record in the plan or its data is a blocking finding — that data is synthetic by specification, and this review is the only gate it passes through.
  5. Conventions: every entry in the profile's conventions and review_checklist; tests present and meaningful (not tautological).
  6. Duplication: did this duplicate something that should have been extended? Check the profile's reuse_hotspots.

Read the full file on GitHub · 67 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 Changed · +6 lines 113bd269cbc2
  2. 10d ago First seen · 61 lines · 14 tokens per session scan A 732f993af358

Subscribe to this mod's changes

reviewer is a command published in the GitHub repository genai-jerry/claude-software-factory (2 stars, last pushed today), licensed MIT. It adds 14 tokens to every session and 866 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 2026-08-31.