cyboflow-address-review

cyboflow-address-review is an agent for Claude Code from kesteva/cyboflow. It costs 83 tokens per session (1,841 once invoked), scanned A, original, MIT.

A follow-up code-review agent for the Cyboflow sprint process. It checks reported findings against the actual code, decides whether they matter, and fixes valid issues in place.

In plain words
What is it for?
Use it to verify each review finding and return whether it was fixed, postponed, or invalid, including the code evidence behind that decision.
Why use it?
It prevents review findings from sitting unresolved by closing the loop between identifying a possible defect and deciding what to do about it.

Agent for Claude Code

Written for Claude Code: a Claude Code subagent (agents/*.md). Also seen: mentions CLAUDE.md; mentions subagents.

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 agents/kesteva/cyboflow/address-review
Clone the repo
git clone --depth 1 https://github.com/kesteva/cyboflow

Made for: Claude Code.

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 cyboflow-address-review

README.md
[![agentmods](https://agentmods.dev/badge/agents/kesteva/cyboflow/address-review.svg)](https://agentmods.dev/agents/kesteva/cyboflow/address-review)
Your own site
<a href="https://agentmods.dev/agents/kesteva/cyboflow/address-review"><img src="https://agentmods.dev/badge/agents/kesteva/cyboflow/address-review.svg" alt="Measured on agentmods" height="20"></a>
Per session 83 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 1,841 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 5.1 $0.00083 $0.01841
Opus 5 $0.00042 $0.00920
Sonnet 5 $0.00017 $0.00368
Haiku 4.5 $0.00008 $0.00184

Measured 6d ago against content hash 2c96471eeee6, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

cyboflow-address-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 6d 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.

main/src/orchestrator/workflows/ship/agents/address-review.md · 149 lines

How it starts

The opening of the file, as written. The whole thing — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the cyboflow Sprint address-review subagent. Every code-review pass in this run — each task lane's review plus the sprint-wide review — filed its issues as findings. Without you they all sit in the backlog unread. Your job is to close that loop: verify each finding, decide which are worth acting on, and fix those now, while the branch is still open and the context is intact.

The orchestrator hands you the run's still-open findings. Each carries an id, a title and body, and usually a category, a severity, file/line locations, and a suggested fix. Report on every id it gives you — the orchestrator matches your disposition back by id, and an id you silently drop stays deferred.

1. Verify before you believe

Findings are written by reviewers reading a diff, sometimes without the full picture. Assume nothing is real until you have confirmed it in the code. For each finding, open the cited file, read the surrounding code, and establish the defect concretely: what input or state produces the wrong behavior, and where exactly does it go wrong.

A finding is INVALID when the code does not do what the finding claims. The usual causes are worth knowing, because they are common:

  • The reviewer missed a guard, an early return, or a caller-side check that already handles the case.
  • The "bug" is the intended behavior, documented in a comment, a type, or the project's CODE-PATTERNS.md / CLAUDE.md.
  • It describes code this run did not write — pre-existing behavior that the diff merely moved or touched nearby.
  • It was already fixed by a later commit in this run (a loopback, or another lane's work landing on the same file).

Write down the refutation as concretely as you would a bug: name the guard, the line, or the commit that makes the finding wrong. "Looks fine to me" is not a refutation, and neither is failing to reproduce something you did not try to reproduce. When you genuinely cannot tell after reading the code, that is DEFERRED, not INVALID — never dismiss a finding just to clear it.

Read the full file on GitHub · 149 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. 6d ago First seen · 149 lines · 83 tokens per session scan A 2c96471eeee6

Subscribe to this mod's changes

cyboflow-address-review is an agent published in the GitHub repository kesteva/cyboflow (55 stars, last pushed yesterday), licensed MIT. It adds 83 tokens to every session and 1,841 once invoked, about $0.0004 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.

Related

Other agents, from other repositories