coder-eval-code-review

A command that reviews uncommitted code changes with several AI models and applies fixes for high-confidence findings of medium severity or higher. Uncommitted changes are edits not yet recorded in a Git commit.

In plain words
What is it for?
Use it before release to review staged and unstaged edits, or describe a change so the command can locate and inspect the relevant files.
Why use it?
It combines verification and independent reviews to catch issues before changes are shipped, while focusing on the files currently being changed.

Command 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 commands/uipath/coder_eval/coder-eval-code-review
Clone the repo
git clone --depth 1 https://github.com/UiPath/coder_eval

Made for: Claude Code.

Per session 17 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,210 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 $0.00017 $0.02210
Opus 5 $0.00009 $0.01105
Sonnet 5 $0.00003 $0.00442
Haiku 4.5 $0.00002 $0.00221

Measured yesterday against content hash aee8c8373420, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

coder-eval-code-review 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 yesterday.

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.

.claude/commands/coder-eval-code-review.md · 155 lines

How it starts

The opening of the file, as written. The whole thing — 155 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Context

  • Current git status: !git status --short
  • Current branch: !git branch --show-current
  • Uncommitted changes: !git diff --name-only && git diff --name-only --cached

Your Task

Run a thorough code review using multiple AI models in parallel, then fix all high-confidence medium-severity and above findings.

Cost note: This command runs make verify + a multi-model review (gemini-3 + gpt-5 via MCP, plus an Opus sub-agent) + automatic fixes. Expensive in time and tokens. Use this for changes you're about to ship; for broad codebase audits use /coder-eval-code-review-full; for quick local checks, run targeted tools (ruff, pyright, pytest) directly.

Input: $ARGUMENTS

Determining Scope

If input was provided: Interpret it as a description of what was changed. Use it to identify the relevant files — search the codebase, read recent commits, check git status — then read those files in full.

If no input was provided: Review all locally modified files (unstaged + staged, not yet committed). Identify them from git diff --name-only and git diff --name-only --cached. If there are no uncommitted changes, review the files changed in the most recent commit (git diff HEAD~1 --name-only).

Read all identified files in full before launching any review.

Severity Standard

Tag every finding 🔴 Critical / 🟠 High / 🟡 Medium / 🔵 Low. Calibrate by impact if shipped, not by fix difficulty. When torn between two levels, pick the lower one. Use the anchors below to keep ratings reproducible across runs and across reviewers.

🔴 Critical 🟠 High 🟡 Medium 🔵 Low
Data loss; secrets/credentials leak; security hole reachable in production; crash on the golden path; silently-wrong EvaluationResult Wrong behavior on a common path; race condition; resource leak (subprocess, tempdir, file handle); bare except swallowing a specific error class; logic fix landing without a regression test Design issue (KISS/DRY/CLAUDE.md violation); Checklist hit outside a hot path; weak test that wouldn't catch the regression; missing extra="forbid" on a config model Style or naming nit; missing internal docstring; redundant comment; example in --help is stale

Read the full file on GitHub · 155 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. yesterday First seen · 155 lines · 17 tokens per session scan A aee8c8373420

Subscribe to this mod's changes

coder-eval-code-review is a command published in the GitHub repository UiPath/coder_eval (119 stars, last pushed 3d ago), licensed Apache-2.0. It adds 17 tokens to every session and 2,210 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 2026-08-30.