review

review is a command for coding agents from ramantiz/planship. It costs 24 tokens per session (590 once invoked), scanned A, original, MIT.

A command that checks a PLAN.md file, a Markdown document describing phases, tasks, dependencies, and acceptance criteria. It reports structural errors and warnings without changing the plan or repository.

In plain words
What is it for?
Use it to validate plans after editing and to spot missing dependencies, incorrect task order, or completion criteria that people cannot verify.
Why use it?
It catches broken task links, dependency cycles, duplicate names, empty phases, and incomplete acceptance details before the plan is shipped to GitHub.

Command

Installs and runs on its own, but its text points at files inside its plugin — anything it tells you to read at a ${CLAUDE_PLUGIN_ROOT} path is only there once the plugin is installed. Installing the plugin gets both.

Part of the planship plugin — 3 commands shipped together

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/ramantiz/planship/review
Clone the repo
git clone --depth 1 https://github.com/ramantiz/planship

Or install planship, the plugin that ships this one along with the rest of its 3 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 review

README.md
[![agentmods](https://agentmods.dev/badge/commands/ramantiz/planship/review.svg)](https://agentmods.dev/commands/ramantiz/planship/review)
Your own site
<a href="https://agentmods.dev/commands/ramantiz/planship/review"><img src="https://agentmods.dev/badge/commands/ramantiz/planship/review.svg" alt="Measured on agentmods" height="20"></a>
Per session 24 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 590 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 $0.00024 $0.00590
Opus 5 $0.00012 $0.00295
Sonnet 5 $0.00005 $0.00118
Haiku 4.5 $0.00002 $0.00059

Measured 3d ago against content hash bb3c40c473fd, 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 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.

commands/review.md · 52 lines

What it actually says

Validate a planship PLAN.md after a human has edited it. Create nothing. Change nothing — not the plan, not the repo. Your output is a report.

Input: $ARGUMENTS — path to the plan (default PLAN.md).

Steps

  1. Run the deterministic validator:

    node "${CLAUDE_PLUGIN_ROOT}/scripts/parse-plan.mjs" <plan-path> --report
    

    It reports, with line numbers:

    • Errors (these block /planship:ship): a Blocked by naming a task that doesn't exist (with a "did you mean" suggestion), dependency cycles, self-dependencies, duplicate task titles, duplicate phase names, a ### task before any ## phase, an empty plan.
    • Warnings (these do not block): a phase with no tasks, a phase with no > Milestone: line, a task with no Done when:, a task with no Scope:.
  2. Read the plan yourself and add what a parser cannot judge:

    • Missing dependencies — a task that clearly cannot start before another one, with no Blocked by. This is the most valuable thing you can find.
    • Wrong ordering — a task in Phase 0 that depends on Phase 2 work.
    • Unverifiable acceptance — a Done when: that no human could actually check ("it works", "the code is clean").
    • Tasks that are really several tasks — a Scope list covering unrelated work.
    • Phases that aren't milestones — a phase that doesn't leave the project in a checkable state.
  3. Report:

    • The totals (phases, tasks, dependency edges).
    • Every error, with the file:line and a concrete suggested fix (the exact replacement text).
    • Every warning worth acting on.
    • Your own findings from step 2, clearly marked as judgement rather than a hard error.
    • A verdict: ready to ship, or N errors must be fixed first.
  4. Do not apply the fixes. Offer them; the human edits the file. If they then ask you to apply a specific fix, that's a normal edit request and you may do it — but never as part of this command's default flow.

If the validator exits 0 with no warnings, say so plainly and point them at /planship:ship <path>.

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 · 52 lines · 24 tokens per session scan A bb3c40c473fd

Subscribe to this mod's changes

review is a command published in the GitHub repository ramantiz/planship (10 stars, last pushed 1mo ago), licensed MIT. It adds 24 tokens to every session and 590 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.