code-review

An automated pull-request code review that uses several specialized agents. A pull request is a proposed set of changes for merging into a shared codebase.

In plain words
What is it for?
Use it to review a current pull request, check project instructions, inspect its changes, and run several review checks in parallel.
Why use it?
It organizes checks for guidelines, bugs, project history, and whether linked issues or designs were actually addressed.

Command

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/marmot-protocol/agent-config/code-review
Clone the repo
git clone --depth 1 https://github.com/marmot-protocol/agent-config
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 635 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 $0.00011 $0.00635
Opus 5 $0.00005 $0.00318
Sonnet 5 $0.00002 $0.00127
Haiku 4.5 $0.00001 $0.00064

Measured yesterday against content hash c1a9c41bdc84, 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 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.

opencode/commands/code-review.md · 97 lines

How it starts

The opening of the file, as written. The whole thing — 97 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Code Review

Perform automated code review on the current pull request.

Process

Step 1: Gate Check

First, invoke @cr-gatekeeper to check if review is needed. If it returns SKIP_REVIEW, stop and explain why.

Step 2: Gather Guidelines

Invoke @cr-guidelines to find all relevant CLAUDE.md and AGENTS.md files. Note the list for the compliance check.

Step 3: Summarize Changes

Invoke @cr-summarizer to get an overview of the PR changes. This provides context for the review agents.

Step 4: Parallel Review

Launch these review tasks in parallel by invoking ALL of them in a SINGLE response (multiple Task tool calls at once):

  1. @cr-compliance: Check guideline compliance

    • Pass the list of guideline files from Step 2
    • Pass the PR title and description for context
  2. @cr-bugs: Scan for obvious bugs

    • Focus only on the diff
    • Pass the PR title and description for context
  3. @cr-history: Analyze git history

    • Look for context-based issues
    • Pass the PR title and description for context
  4. @cr-issues: Verify linked issue resolution and Figma designs

    • Check if PR actually fixes the issues it claims to close
    • If Figma links are present, verify implementation matches the design
    • Pass the PR number for issue extraction

IMPORTANT: Call all four Task tools in parallel (same message), do not wait for one to complete before starting the next.

Collect all issues from all agents.

Step 5: Validate Issues

For each issue found with confidence < 90:

  • Invoke @cr-validator to independently verify
  • Pass the issue details and PR context
  • Update confidence based on validation

Step 6: Filter

Filter out any issues with adjusted confidence less than 80. These are likely false positives.

Step 7: Report

If issues remain after filtering:

Format each issue as:

## Code Review Issues

Found {N} issues:

### 1. {Issue Title}
**Type**: {compliance|bug|history|issue-resolution}
**Confidence**: {score}/100
**Location**: {file}:{line}

{Description}

{Code snippet with context}

**Suggestion**: {Fix suggestion if applicable}

---

Read the full file on GitHub · 97 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 · 97 lines · 11 tokens per session scan A c1a9c41bdc84

Subscribe to this mod's changes

code-review is a command published in the GitHub repository marmot-protocol/agent-config (2 stars, last pushed 7mo ago), licensed MIT. It adds 11 tokens to every session and 635 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.