A code-review workflow that sends a code change to specialist reviewers, checks their findings, and prepares comments for your approval.

Use it to review diffs for logic, architecture, machine learning, statistics, data engineering, prompt engineering, and related concerns.

It helps catch problems that a single review may miss, while filtering unsupported or low-quality findings before they reach the code review.

code-review — Skill for Claude Code and Codex by kbichave: A code-review workflow that sends a code change to…

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 skills/kbichave/skills/code-review

Any agent

npx skills add kbichave/skills --skill code-review

Clone the repo

git clone --depth 1 https://github.com/kbichave/skills

Made for: Claude Code, Codex.

Or install deep, the plugin that ships this one along with the rest of its 4 skills, 16 agents, 6 hooks.

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/skills/kbichave/skills/code-review.svg)](https://agentmods.dev/skills/kbichave/skills/code-review)

Your own site

<a href="https://agentmods.dev/skills/kbichave/skills/code-review"><img src="https://agentmods.dev/badge/skills/kbichave/skills/code-review.svg" alt="Measured on agentmods" height="20"></a>

Per session 217 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 6,220 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.

Model	Per session	Once invoked
Fable 5	$0.00217	$0.06220
Opus 5	$0.00109	$0.03110
Sonnet 5	$0.00043	$0.01244
Haiku 4.5	$0.00022	$0.00622

Measured 3d ago against content hash 0aeed6d4b053, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

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 3d 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.

skills/code-review/SKILL.md · 500 lines

How it starts

The opening of the file, as written. The whole thing — 500 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Code Review (standalone, expert panel)

Multi-expert review: the diff decides which specialist subagents spawn, two verifier stages filter their findings, and only user-approved comments land in the source.

No-trace rule (overrides everything below)

Nothing this skill externalizes may reveal that AI, a skill, a plugin, an agent panel, or a model produced it. The output is the user's own review.

Commits made during the flow: plain messages only. NEVER add Co-Authored-By: Claude, Generated with Claude Code, 🤖, or any tool/model/skill mention. (This overrides any default commit trailer.)
PR body + comments: no attribution, no panel/verifier stats, no expert names, no rule_id/tag codes, no "Automated/AI/generated" wording.
Marker text in source: neutral CODECHANGE(review): / RECOMMENDATION(review): only — no AI/skill/plugin/agent words.
Report file: written outside the repo (see step 7), so it cannot be staged, committed, or need a .gitignore entry. It may name experts/packs for the user's own use; it must never be copied into the repo or any shared surface.
If the user later commits the reviewed code, the diff carries no evidence a tool touched it.

Flow

1. Scope

Determine what to review:

User named a base (main, a commit, a tag): git diff <base>...HEAD --name-only.
User named files: use those.
Nothing named: default to uncommitted + staged changes (git diff HEAD --name-only); if clean, ask for a base.

Fail fast before any agent spawns: git rev-parse <base> must resolve and the diff must be non-empty. A bad ref or empty diff dies here, not inside a sub-agent.

Size guardrail. Measure the diff (git diff <base> --shortstat). Defect detection collapses past ~400 changed LOC (SmartBear). If the diff exceeds ~400 LOC, tell the user, and offer to either (a) scope the review to a subset of files/commits now, or (b) proceed but run in paced passes so no region gets a shallow read. Proceed whole only on the user's say-so.

Read the full file on GitHub · 500 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.

3d ago First seen · 500 lines · 0 tokens per session scan A 0aeed6d4b053

Subscribe to this mod's changes

code-review is a skill published in the GitHub repository kbichave/skills (2 stars, last pushed 4d ago), licensed MIT. It adds 217 tokens to every session and 6,220 once invoked, about $0.0011 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.

Code review ai-planning claude-code claude-code-plugin implementation planning multi-llm

council

Convene the Council of High Intelligence — multi-persona deliberation with historical thinkers for deeper analysis of complex problems.

0xNyk/council-of-high-intelligence · 27 tokens

map-plan

ARCHITECT phase - decompose complex tasks into atomic subtasks with research, spec, and branch-scoped plan artifacts under .map.

azalio/map-framework · 31 tokens

map-tdd

TDD MAP workflow: write tests from the spec FIRST, then implement, so tests validate intent not implementation. Use when correctness is critical (auth, payments, data integrity). Do NOT use without a spec; use map-efficient instead.

azalio/map-framework · 51 tokens

map-review

Interactive 4-section code review using monitor, predictor, and evaluator agents plus the user and maintainer role reviewers on current changes. Use when reviewing a diff, PR, or staged work before merge. Do NOT use to plan or implement; use map-plan or map-efficient.

azalio/map-framework · 58 tokens

map-debug

Structured MAP debugging via task-decomposer, actor, and monitor agents. Use when reproducing a bug, isolating a regression, or diagnosing an error with specialized agents — including failing or flaky tests (pytest AssertionError), crashes and segmentation faults, memory-corruption or memory errors in native/C…

azalio/map-framework · 213 tokens

map-learn

Capture reusable lessons after a completed MAP workflow. Use when a MAP run has finished and you want rules written to .claude/rules/learned/ from a workflow summary or handoff. Do NOT use during active implementation.