fd-review

fd-review is a command for coding agents from DVNghiem/FlowDeck. It costs 29 tokens per session (1,391 once invoked), scanned A, original, MIT.

A pre-coding review of a planned task from two viewpoints: one checks whether the work is worth doing and correctly scoped, while the other checks the technical design and possible side effects.

In plain words
What is it for?
Use it to review task requirements, architecture, affected files, risks, and implementation steps before writing code.
Why use it?
It catches unclear requirements, design problems, missing edge cases, and affected code areas before implementation begins.

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/dvnghiem/flowdeck/fd-review
Clone the repo
git clone --depth 1 https://github.com/DVNghiem/FlowDeck

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 fd-review

README.md
[![agentmods](https://agentmods.dev/badge/commands/dvnghiem/flowdeck/fd-review.svg)](https://agentmods.dev/commands/dvnghiem/flowdeck/fd-review)
Your own site
<a href="https://agentmods.dev/commands/dvnghiem/flowdeck/fd-review"><img src="https://agentmods.dev/badge/commands/dvnghiem/flowdeck/fd-review.svg" alt="Measured on agentmods" height="20"></a>
Per session 29 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,391 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 $0.00029 $0.01391
Opus 5 $0.00015 $0.00696
Sonnet 5 $0.00006 $0.00278
Haiku 4.5 $0.00003 $0.00139

Measured 4d ago against content hash baf3bc5ae869, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

fd-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 4d 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.

docs/commands/fd-review.md · 168 lines

How it starts

The opening of the file, as written. The whole thing — 168 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Review

Review the artifacts /fd-task produced, through two independent lenses, before any code is written.

Input: $ARGUMENTS — optional --topic=<slug> to review a topic other than the active one.

Step 1: Load artifacts

Resolve <topic> from --topic, else from topic in ~/.fd-plan/<slug>/STATE.md.

MUST validate artifacts before review:

Call fdx-validate action:artifacts topic:<topic>.

If valid: false → STOP. Print:

❌ Artifact validation failed. Fix these issues before fd-review:
  <error 1>
  <error 2>
Run /fd-task to regenerate the artifacts.

If valid: true → proceed to load artifacts.

Read all four from ~/.fd-plan/<slug>/<topic>/:

  • task.md — requirements, acceptance criteria, constraints
  • architecture.md — the proposed design
  • affect.md — affected files, risk level, parallel safety
  • plan.md — implementation steps and waves

If any is missing:

Error: <file> not found for topic "<topic>". Run /fd-task first.

Also read ~/.fd-plan/<slug>/architecture.md for project-level context.

Also call: repo-memory action:search query:"review lessons " The agent MUST load any prior review findings into context before starting CEO or eng review.

Step 2: CEO review — challenge the premise

Argue with the task, not the code. Answer each question explicitly:

  1. Is this the right problem? Does solving it move anything that matters, or is it work for its own sake? What happens if we do nothing?
  2. Is this the right approach? Is there a materially cheaper or simpler way to get the same outcome? Is any part of the scope speculative?
  3. Is the risk acceptable? Given the risk level in affect.md, is the payoff worth the blast radius? What is the worst realistic outcome?
  4. Is the scope right? What in task.md could be cut without losing the point? What is missing that would make this fail on delivery?

Be direct. If the premise is weak, say so plainly rather than hedging.

Read the full file on GitHub · 168 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. 4d ago First seen · 168 lines · 29 tokens per session scan A baf3bc5ae869

Subscribe to this mod's changes

fd-review is a command published in the GitHub repository DVNghiem/FlowDeck (24 stars, last pushed 15d ago), licensed MIT. It adds 29 tokens to every session and 1,391 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.