octo-review

octo-review is a command for Claude Code from CES-Ltd/Lumi. It costs 18 tokens per session (1,642 once invoked), scanned A, original, MIT.

A command for reviewing code by sending it through several large language models and adding comments directly to a pull request. A pull request is a proposed change to a codebase for review before it is merged.

In plain words
What is it for?
Running multi-model reviews of the scope chosen by the user and producing inline pull-request comments.
Why use it?
It gathers review opinions from multiple AI systems instead of relying on one review, while keeping findings attached to the proposed code changes.

Command for Claude Code

Written for Claude Code: shipped in a Claude Code plugin. Also seen: mentions subagents; names the AskUserQuestion tool; mentions Claude Code.

Part of the octo plugin — 10 skills, 47 commands shipped together

Good fit Running multi-model reviews of the scope chosen by the user and producing inline pull-request comments.

Compare 6 commands from other repositories ↓
Install with agentmods
npx agentmods add commands/ces-ltd/lumi/octo-review
View source ↗ CES-Ltd/Lumi
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/CES-Ltd/Lumi

Made for: Claude Code.

Or install octo, the plugin that ships this one along with the rest of its 10 skills, 47 commands.

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 octo-review

README.md
[![agentmods](https://agentmods.dev/badge/commands/ces-ltd/lumi/octo-review/github.svg)](https://agentmods.dev/commands/ces-ltd/lumi/octo-review)
Your own site
<a href="https://agentmods.dev/commands/ces-ltd/lumi/octo-review"><img src="https://agentmods.dev/badge/commands/ces-ltd/lumi/octo-review/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 octo-review

Your own site · 80×15
<a href="https://agentmods.dev/commands/ces-ltd/lumi/octo-review"><img src="https://agentmods.dev/badge/commands/ces-ltd/lumi/octo-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 18 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,642 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 $0.00018 $0.01642
Opus 5 $0.00009 $0.00821
Sonnet 5 $0.00004 $0.00328
Haiku 4.5 $0.00002 $0.00164

Measured 11d ago against content hash 2b8f64e31ad3, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

octo-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 11d 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.

resources/seed/octopus/commands/octo-review.md · 167 lines

How it starts

The opening of the file, as written. The whole thing — 167 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/octo:review

MANDATORY COMPLIANCE — DO NOT SKIP

When the user invokes /octo:review, you MUST execute the multi-LLM review workflow below. You are PROHIBITED from:

  • Doing a direct code review without running the structured workflow
  • Deciding the scope is "too broad" and narrowing it yourself — the user chose the scope
  • Skipping the provider check or AskUserQuestion steps
  • Substituting a simpler review approach because it seems "more effective"
  • Running two background Sonnet agents instead of the full multi-provider pipeline

The user chose /octo:review over a regular code review deliberately. They want multi-provider perspectives (Codex + Gemini + Claude), not a single-model review. If you catch yourself thinking "a focused audit would be more effective" — STOP. That is the exact rationalization this instruction prohibits.

EXECUTION MECHANISM — NON-NEGOTIABLE

You MUST execute this command by calling orchestrate.sh as documented below. You are PROHIBITED from:

  • ❌ Doing the work yourself using only Claude-native tools (Agent, Read, Grep, Write)
  • ❌ Using a single Claude subagent instead of multi-provider dispatch via orchestrate.sh
  • ❌ Skipping orchestrate.sh because "I can do this faster directly"

Multi-LLM orchestration is the purpose of this command. If you execute using only Claude, you've violated the command's contract.


🐙 CLAUDE OCTOPUS ACTIVATED — Multi-LLM Code Review

Providers: 🔴 Codex CLI — logic and correctness 🟡 Gemini CLI — security and edge cases 🔵 Claude — architecture and synthesis 🟣 Perplexity — CVE lookup (if available)


When the user invokes this command (e.g., /octo:review <arguments>):

Step 1: Ask Clarifying Questions / Context Acquisition

Determine mode based on session autonomy:

If AUTONOMY_MODE env var is autonomous, or session is running headlessly, or OCTOPUS_WORKFLOW_PHASE is set (indicating a pipeline context like /octo:develop or /octo:embrace), skip Q&A and auto-infer with ALL focus areas:

  1. Run git diff --cached — if non-empty, target=staged
  2. Run gh pr view --json number — if open PR exists, set target=<pr_number>
  3. Otherwise target=working-tree
  4. Set provenance=unknown, autonomy=autonomous, publish=ask, debate=auto, focus=["correctness","security","architecture","tdd"]

Read the full file on GitHub · 167 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. 11d ago First seen · 167 lines · 18 tokens per session scan A 2b8f64e31ad3

Subscribe to this mod's changes

octo-review is a command published in the GitHub repository CES-Ltd/Lumi (31 stars, last pushed 4mo ago), licensed MIT. It adds 18 tokens to every session and 1,642 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.