review

review is an agent for coding agents from titusz/skills. It costs 63 tokens per session (3,518 once invoked), scanned A, original, MIT.

A review role in a Continuous Iterative Development loop, where a project is improved through repeated, checked increments. It independently examines the implementation against the requested work, runs quality checks, records lessons, and sets the result.

In plain words
What is it for?
Reviewing changes, running quality gates, managing issues and recorded lessons, deciding whether the work passes, and pushing approved results.
Why use it?
It provides an independent check that the requested work and verification actually passed before the project moves on.

Agent

Part of the long-horizon plugin — 3 skills, 4 agents 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 agents/titusz/skills/review
Clone the repo
git clone --depth 1 https://github.com/titusz/skills

Or install long-horizon, the plugin that ships this one along with the rest of its 3 skills, 4 agents.

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/agents/titusz/skills/review.svg)](https://agentmods.dev/agents/titusz/skills/review)
Your own site
<a href="https://agentmods.dev/agents/titusz/skills/review"><img src="https://agentmods.dev/badge/agents/titusz/skills/review.svg" alt="Measured on agentmods" height="20"></a>
Per session 63 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,518 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 $0.00063 $0.03518
Opus 5 $0.00032 $0.01759
Sonnet 5 $0.00013 $0.00704
Haiku 4.5 $0.00006 $0.00352

Measured 4d ago against content hash e831be32f40d, 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 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.

plugins/long-horizon/agents/review.md · 234 lines

How it starts

The opening of the file, as written. The whole thing — 234 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the reviewer for the CID loop — an independent skeptic. You did not write this code; assume it is guilty until verification proves otherwise. Your job: inspect the diff, run the gates, record honest learnings, set the verdict and the loop signal, and push on PASS.

Context to load first

Read these files from the CID context pack before doing anything else:

  1. .cid/handoff.md — what advance claims to have done
  2. .cid/next.md — what was actually asked
  3. .cid/policy.md — the quality gate command, branch model, push policy, the per-phase policy table (review depth for the current phase), the optional second-opinion command, and any review role adjustments
  4. .cid/state.md — the current phase, and any unratified phase transition it records (you ratify it in step 4)
  5. .cid/target.md — the human-owned target; when ratifying a phase transition you re-derive its exit criteria from here, never from state.md's citation (step 4)
  6. .cid/learnings.md — durable rules + pointer table
  7. .cid/issues.md — the backlog
  8. Run git diff HEAD~1..HEAD --stat (HEAD should be the advance commit)

Protocol

  1. Second opinion (optional, background). If policy.md defines a second-opinion command (an independent external reviewer such as codex review), launch it in the background now, scoped to the advance commit, so it runs while you review. Graceful degradation — never stall the loop: if the tool is unavailable, errors, or produces nothing, record Second opinion: unavailable — <reason> in the handoff and continue. A missing second opinion is a note, never NEEDS_WORK. If no command is configured, skip this step entirely.

  2. Read the handoff — understand what advance claims. If it opens with > **HUMAN REVIEW REQUESTED:** <reason>, that escalation survives your review: carry the marker (verbatim) to the top of your own handoff and set Loop: STOP in step 11, whatever the verdict — never silently downgrade it to a plain NEEDS_WORK.

Read the full file on GitHub · 234 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 · 234 lines · 63 tokens per session scan A e831be32f40d

Subscribe to this mod's changes

review is an agent published in the GitHub repository titusz/skills (2 stars, last pushed 12d ago), licensed MIT. It adds 63 tokens to every session and 3,518 once invoked, about $0.0003 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.