What is review-assistant?

An interactive tool for reviewing a GitHub pull request, which is a proposed set of code changes. It can walk through findings with you or prepare a review for you to finalize on GitHub.

Use it with a pull-request number to review changes in the current or a specified repository. An automatic mode prepares all qualifying findings as a pending GitHub review.

It organizes pull-request review and lets you set a confidence threshold so that only findings meeting your chosen level are included.

review-assistant — Command for Claude Code by ahardin13

Install

Getting it into your agent

This one installs as part of its plugin. Adding the marketplace and installing the plugin brings it with everything else the plugin ships.

Claude Code

/plugin marketplace add ahardin13/review-assistant

Claude Code

/plugin install review-assistant

Made for: Claude Code.

Or install review-assistant, the plugin that ships this one along with the rest of its 3 skills, 1 command, 1 agent.

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.

README.md

[![agentmods](https://agentmods.dev/badge/commands/ahardin13/review-assistant/review-assistant/github.svg)](https://agentmods.dev/commands/ahardin13/review-assistant/review-assistant)

Your own site

<a href="https://agentmods.dev/commands/ahardin13/review-assistant/review-assistant"><img src="https://agentmods.dev/badge/commands/ahardin13/review-assistant/review-assistant/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.

Your own site · 80×15

<a href="https://agentmods.dev/commands/ahardin13/review-assistant/review-assistant"><img src="https://agentmods.dev/badge/commands/ahardin13/review-assistant/review-assistant.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 11 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.

When invoked 980 The whole file, excluding the scripts and references it only reads on demand.

Security scan B 1 finding. 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.

Model	Per session	Once invoked
Fable 5.1	$0.00011	$0.00980
Opus 5	$0.00005	$0.00490
Sonnet 5	$0.00002	$0.00196
Haiku 4.5	$0.00001	$0.00098

Measured 9d ago against content hash 8a9fc0301483, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-09, from the pricing page.

Security

Grade B, and why

review-assistant 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 9d 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.

allowed-tools: Bash(gh pr view:*), Bash(gh pr diff:*), Bash(gh pr comment:*), Bash(gh pr review:*), Bash(gh pr list:*), Bash(gh issue view:*), Bash(gh api:*), Bash(git remote:*), Bash(git log:*), Bash(git blame:*), Bash(

commands/review-assistant.md · 51 lines

How it starts

The opening of the file, as written. The whole thing — 51 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Review a pull request interactively.

Usage:

/review-assistant <PR_NUMBER> — review PR in current repo
/review-assistant <PR_NUMBER> --repo owner/repo — review PR in specified repo
/review-assistant <PR_NUMBER> --threshold N — set minimum confidence score (default: 50)
/review-assistant <PR_NUMBER> --auto — skip the interactive walkthrough and post all findings as a pending GitHub review for the user to finalize on github.com

If no PR number is provided, show usage and exit.

You are a dispatcher. Do NOT review code, analyze diffs, present findings, or interact with the user about review content yourself. Do NOT pause, summarize, or wait for user input between steps — proceed immediately through all steps.

1. Parse inputs

pr_number: required
repo: optional (default: parse from git remote get-url origin)
threshold: optional confidence threshold (default: 50)
auto: boolean, true if --auto present (default: false)

If no repo can be determined, exit: "Run from within the repo, or pass --repo owner/repo."

2. Announce mode

This must be your first user-facing line. Exactly one of:

If auto is true: "Running auto review — findings will be posted as a pending review on PR #<N> for you to finalize on GitHub."
Otherwise: "Starting interactive walkthrough — I'll step through PR #<N> with you file by file."

3. Gather context and run analysis

Call: Skill("review-assistant:reading-pr-context", args: "pr_number=<N> repo=<owner/repo> threshold=<T>")

This skill handles PR metadata, diff, eligibility checks, session file creation, REVIEW.md loading, code-review analysis, and writing findings to the session file. The session file path will be the most recent file matching $HOME/.local/state/review-assistant/sessions/pr-<N>-*.md.

Do not stop here. This step ends in the middle of the dispatch flow — Step 4 has not yet run. Do not print "session file ready", "returning to orchestrator", or any other closing text. Your next tool call MUST be the Skill dispatch in Step 4.

Read the full file on GitHub · 51 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.

9d ago First seen · 51 lines · 11 tokens per session scan B 8a9fc0301483

Subscribe to this mod's changes

review-assistant is a command published in the GitHub repository ahardin13/review-assistant (3 stars, last pushed 3mo ago), licensed MIT. It adds 11 tokens to every session and 980 once invoked, about $0.0001 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 2026-08-31.

Claude Code Code review Git and workflow claude-code-plugin pull-request

review-assistant