team-review

team-review is a command for coding agents from dshakes/compass. It costs 41 tokens per session (408 once invoked), scanned A, original, MIT.

A command that reviews a code change with several agents working in parallel and sharing their findings. It checks correctness, security, and tests, then combines the results into one report.

In plain words
What is it for?
Use it to review a Git diff or pull request before merging. It is intended for interactive reviews and requires an experimental agent-team setting.
Why use it?
It reduces the chance that a single review misses an important problem. It also avoids waiting for reviewers to work one after another.

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/dshakes/compass/team-review
Clone the repo
git clone --depth 1 https://github.com/dshakes/compass

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 team-review

README.md
[![agentmods](https://agentmods.dev/badge/commands/dshakes/compass/team-review.svg)](https://agentmods.dev/commands/dshakes/compass/team-review)
Your own site
<a href="https://agentmods.dev/commands/dshakes/compass/team-review"><img src="https://agentmods.dev/badge/commands/dshakes/compass/team-review.svg" alt="Measured on agentmods" height="20"></a>
Per session 41 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 408 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 $0.00041 $0.00408
Opus 5 $0.00020 $0.00204
Sonnet 5 $0.00008 $0.00082
Haiku 4.5 $0.00004 $0.00041

Measured 3d ago against content hash dfb5b62ef9b0, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

team-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.

claude/commands/team-review.md · 27 lines

What it actually says

Review ${ARGUMENTS:+$ARGUMENTS }with a coordinated agent team (parallel, not sequential).

Experimental. Requires CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1 (set in this repo's settings.json). Agent teams have rough edges (no teammate resumption, task-status lag). Prefer this for interactive review; the headless pipeline stays sequential.

Spin up a small team and have them message each other, not just report in isolation:

  1. Lead (you) — gather the diff (git diff / gh pr view), create a shared task list, and spawn three teammates with disjoint mandates:
    • Correctness — logic, edge cases, error paths, conventions (model: sonnet)
    • Security — authz, injection, secrets, tenancy, trust boundaries (model: opus)
    • Tests/QA — coverage gaps, missing regression tests, flake risk (model: haiku/sonnet)
  2. Cross-talk — teammates SendMessage when findings interact (e.g., Security flags an input that Correctness approved; Tests notes the fix lacks coverage). Resolve conflicts between them, don't just concatenate.
  3. Lead synthesis — produce ONE reconciled report grouped Blocking / Should-fix / Nit, noting where reviewers agreed vs. disagreed and how you resolved it.

Do not edit files or merge — this is review only. The human still owns the merge gate.

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. 3d ago First seen · 27 lines · 41 tokens per session scan A dfb5b62ef9b0

Subscribe to this mod's changes

team-review is a command published in the GitHub repository dshakes/compass (19 stars, last pushed 10d ago), licensed MIT. It adds 41 tokens to every session and 408 once invoked, about $0.0002 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.