Ships with octo. Installing the plugin gets this command.
⚡ How it fires
How this command gets triggered: by you, by Claude, or both.
Fires itselfClaude auto-loads it when your prompt matches the work.
You can call itInvoke it directly when you want it.
Slash command/staged-review
👁️ Context preview
What this command does when you run it.
[advanced] Two-stage review: spec compliance then code quality
📊 Stats
Stars3,869
Forks362
LanguageShell
LicenseMIT
📦 Ships with octo
</> Command definition
staged-review.md
---command: staged-review
description: "[advanced] Two-stage review: spec compliance then code quality"
aliases:
- two-stage-review
- full-review
---# Staged Review — Two-Stage Pipeline
## 🤖 INSTRUCTIONS FOR CLAUDE
### MANDATORY COMPLIANCE — DO NOT SKIP
**When the user explicitly invokes `/octo:staged-review`, you MUST execute the structured workflow below.** You are PROHIBITED from doing the task directly, skipping the review phases, or deciding the task is "too simple" for this workflow. The user chose this command deliberately — respect that choice.
### EXECUTION MECHANISM — NON-NEGOTIABLE
**You MUST dispatch work to external providers (Codex, Gemini, Antigravity, etc.) for this command. You are PROHIBITED from:**
- ❌ Executing the entire task using only Claude-native tools
- ❌ Using a single Agent subagent instead of multi-provider dispatch
- ❌ Skipping provider dispatch because "I can handle this alone"
**Multi-LLM orchestration is the purpose of this command.** Single-model execution defeats its purpose.
---🐙 **CLAUDE OCTOPUS ACTIVATED** — Two-Stage Review Pipeline
---When the user invokes this command (e.g., `/octo:staged-review` or `/octo:staged-review src/auth/`):
**Before Step 1, check for governed mode.** If a `.todo/claude-octopus/` directory exists in the working tree (set up by the Cruisey `octo` pipeline), this review is running INSIDE the autonomous governed workflow — the human gate is CP3, not here. In that case:
- **Skip the Step 1 "Review Scope" prompt** — the scope is always the full working tree (nothing is committed before CP3). Proceed directly with `git diff` of the working tree.
- **Skip the "Post-Completion — Interactive Next Steps" prompt** — proceed straight to the next phase (CP2 council, or the gatekeeper) without asking.
```bash
test -d .todo/claude-octopus && echo "GOVERNED_MODE" || echo "STANDALONE_MODE"
```
If `STANDALONE_MODE`, run the interactive prompts below as normal. If `GOVERNED_MODE`, do NOT call `AskUserQuestion` at any point in this command.
## Step 1: Determine Review Scope
**(GOVERNED_MODE: skip this prompt — scope = working tree.)** Standalone only:
```javascript
AskUserQuestion({
questions: [
{
question: "What should be reviewed?",
header: "Review Scope",
multiSelect: false,
options: [
{label: "Staged changes", description: "git diff --cached — what you're about to commit"},
{label: "Recent commit", description: "Changes in the last commit (HEAD~1..HEAD)"},
{label: "Working tree", description: "All uncommitted changes"},
{label: "Specific path", description: "A file or directory (provide path as argument)"}
]
}
]
})
```
If invoked with arguments (e.g., `/octo:staged-review src/auth/`), use that path directly and skip the question.
1. **Stage 1** (Spec Compliance) — Validate against intent contract
2. **Gate check** — Stage 1 must pass before Stage 2
3. **Stage 2** (Code Quality) — Stub detection + multi-LLM quality review (Codex for logic, Gemini for security, Claude for architecture — synthesized into unified findings)
4. **Combined report** — Unified verdict
### Post-Completion — Interactive Next Steps
**GOVERNED_MODE: do NOT ask — proceed straight to the next phase (CP2 council, then gatekeeper) per the governed pipeline. The next-steps prompt below is STANDALONE-only.**
**CRITICAL (standalone only): After the skill completes, you MUST ask the user what to do next.**
```javascript
AskUserQuestion({
questions: [
{
question: "Review complete. What would you like to do?",
header: "Next Steps",
multiSelect: false,
options: [
{label: "Fix blocking issues", description: "Address the issues flagged in the review"},
{label: "Run /octo:review", description: "Follow up with full multi-LLM review"},
{label: "Post to PR", description: "Post the review findings to the open PR"},
{label: "Done", description: "No further action needed"}
- `/octo:deliver` — Full delivery phase with validation and testing
- `/octo:develop` — Development phase with quality gates
## Validation Gates
- Review scope determined (from arguments or user input)
- Skill loaded and executed (not simulated)
- Combined report presented to user
- Next steps offered
### Prohibited Actions
- Simulating the review without running the skill
- Skipping Stage 1 when an intent contract exists
- Ending without presenting next steps **(standalone mode only — in GOVERNED_MODE the next-steps prompt is intentionally skipped; proceed to the next pipeline phase instead)**
