review

A code-review command that examines changes, checks them against the intended behavior, and assigns confidence to its findings.

In plain words
What is it for?
Use it to review backend, frontend, and security changes, compare code with feature scenarios, and record review findings.
Why use it?
It helps separate likely problems from uncertain observations and keeps review decisions tied to the requested outcome.

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/galando/temper/review
Clone the repo
git clone --depth 1 https://github.com/galando/temper
Per session 13 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 683 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 $0.00013 $0.00683
Opus 5 $0.00006 $0.00342
Sonnet 5 $0.00003 $0.00137
Haiku 4.5 $0.00001 $0.00068

Measured 2d ago against content hash 609d47da7ce9, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

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 2d 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.

commands/review.md · 60 lines

How it starts

The opening of the file, as written. The whole thing — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Review: Confidence-Scored Code Review

Goal: Review changes with parallel subagents, confidence scoring, and intent validation.

Execution

Full methodology: Read $CLAUDE_PLUGIN_ROOT/reference/review.md

Subprocess Mode

If $CLAUDE_PLUGIN_ROOT/scripts/temper config get stages.subprocess false returns true, don't run the methodology inline (skip the reference read and Quick Reference below). Launch the same isolated subprocess /temper uses — model from temper model review, prompt: "Follow $CLAUDE_PLUGIN_ROOT/agents/review.md exactly. Spec: .temper/specs/{feature-slug}. Standalone run — pass --spec-path .temper/specs/{feature-slug} to every temper gate call." Print the returned box verbatim, then run the gate + AskUserQuestion per Deterministic Gate below (the subprocess already recorded each open finding as evidence) — the human gate stays in this context either way.

Quick Reference

  1. Gather changed files + active pack rules + review memory
  2. Launch parallel review subagents (backend/frontend/security)
  3. Structured intent validation: mechanical checks (scenario/code) + deferred (metric/manual)
  4. Filter by confidence threshold + review memory
  5. Generate report to .temper/reviews/
  6. Auto-fix high-priority issues (if enabled, max 2 loops)
  7. Update metrics + review memory
  8. Run $CLAUDE_PLUGIN_ROOT/scripts/temper gate review --spec-path .temper/specs/{feature-slug} and show its PASS/FAIL via AskUserQuestion — see Deterministic Gate below

Active Skills

  • Context Engineering — load hierarchical context at stage start (rules → arch → source → errors, under 2K lines/task)
  • Temper Core — stack detection, pack resolution, quality gates

Diff-aware: focuses on what changed, catches N+1 and performance issues

Deterministic Gate

This is the same gate the unified /temper command's Review stage runs — running this command standalone must not skip it, or temper gate commit sees no review evidence and wrongly blocks (or wrongly passes) a later commit. Follow $CLAUDE_PLUGIN_ROOT/agents/review.md steps 2-3 (record each open finding via temper evidence add --stage review --severity ...) as you review, then run $CLAUDE_PLUGIN_ROOT/scripts/temper gate review and show its PASS/FAIL to the user via AskUserQuestion (this command is not a subprocess — you own the gate here, unlike agents/review.md's "never show a gate" rule).

Read the full file on GitHub · 60 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. 2d ago First seen · 60 lines · 13 tokens per session scan A 609d47da7ce9

Subscribe to this mod's changes

review is a command published in the GitHub repository galando/temper (15 stars, last pushed 3d ago), licensed MIT. It adds 13 tokens to every session and 683 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.