Claude-Mind: Skill for Claude Code

.claude/skills/close-review/SKILL.md

close-review is a skill for Claude Code from zkysar1/Claude-Mind. It costs 123 tokens per session (2,300 once invoked), scanned A, original, MIT.

An independent reviewer for checking whether a completed project goal is ready to close. It compares the goal and its artifacts with the requirements, sources, acceptance criteria, and possible side effects.

In plain words
What is it for?
Use it as a final review step for higher-risk goals before they are marked complete.
Why use it?
It reduces the chance that the person who did the work approves their own mistakes or overlooks missing requirements.

Skill for Claude Code

Written for Claude Code: user-invocable in frontmatter. Also seen: mentions subagents.

This is zkysar1/Claude-Mind's own configuration. It tells Claude Code how to work on Claude-Mind itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything Claude-Mind configures →

Needs its repository: it runs a file that does not travel with it, so clone the repository first. The line is Bash: bash core/scripts/liveness-check.sh --agent <candidate-peer> --json.

Reuse

Borrowing it

Nothing to install: this file belongs to zkysar1/Claude-Mind. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file
curl -O https://raw.githubusercontent.com/zkysar1/Claude-Mind/main/.claude/skills/close-review/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/zkysar1/Claude-Mind

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 close-review

README.md
[![agentmods](https://agentmods.dev/badge/skills/zkysar1/claude-mind/close-review/github.svg)](https://agentmods.dev/skills/zkysar1/claude-mind/close-review)
Your own site
<a href="https://agentmods.dev/skills/zkysar1/claude-mind/close-review"><img src="https://agentmods.dev/badge/skills/zkysar1/claude-mind/close-review/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

agentmods 80×15 button for close-review

Your own site · 80×15
<a href="https://agentmods.dev/skills/zkysar1/claude-mind/close-review"><img src="https://agentmods.dev/badge/skills/zkysar1/claude-mind/close-review.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 123 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,300 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.
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 $0.00123 $0.02300
Opus 5 $0.00062 $0.01150
Sonnet 5 $0.00025 $0.00460
Haiku 4.5 $0.00012 $0.00230

Measured 2d ago against content hash bafd59ca657e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

close-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 2d 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/skills/close-review/SKILL.md · 188 lines

How it starts

The opening of the file, as written. The whole thing — 188 lines — stays where its author put it; the contents beside it link to each section on GitHub.

/close-review — Adversarial Independent Close-Time Review

The verdict-PRODUCER half of the blocking close-review gate. The consumer (close-review-gate.py) refuses a tier-2 close without an APPROVE verdict at the goal-keyed path it owns. This skill is what writes that verdict.

The One Rule That Makes This Worth Running

The reviewer never sees the executing context. It receives the goal record, the artifact paths, and the mandate below — nothing else. A reviewer handed the executor's reasoning inherits the executor's blind spots and re-derives the executor's conclusion, which is a rubber stamp with extra steps. If you cannot run this without the executing context in view, escalate to a fresh-context reviewer per Step 1 instead of proceeding.

Step 0: Load Conventions

Bash: load-conventions.sh with each name from the conventions: front matter.

Step 1: Establish Independence

Independence is the GATE's definition, not this skill's — the producer imports independence_defect rather than reimplementing it, so a self-review is refused at production time (test_self_review_is_REFUSED_at_production_time). Choose the reviewer before assembling anything:

Bash: bash core/scripts/liveness-check.sh --agent <candidate-peer> --json
  • alive → MIND-level review. Route through the existing review-request lane per coordination.md Review Gate. A peer mind carries independent priors and so also catches shared-prior blind spots.
  • dormant / unknown / retired, or no peer exists → fresh-context subagent fallback. Always available, including on solo deployments. Session-level fresh eyes is sufficient for the whole did-not-re-read-the- source class: the founding incident was refutable from the goal description text alone.

Never conclude a peer is unavailable from a stale last_active alone (check-team-state-before-silent.md rule 5) — unknown is not dormant.

Step 2: Assemble the Reviewer's Packet

Exactly three things, and nothing more:

Read the full file on GitHub · 188 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. 2d ago Changed · +42 lines bafd59ca657e
  2. 7d ago First seen · 146 lines · 123 tokens per session scan A d42d256ff184

Subscribe to this mod's changes

close-review is a skill published in the GitHub repository zkysar1/Claude-Mind (5 stars, last pushed 2d ago), licensed MIT. It adds 123 tokens to every session and 2,300 once invoked, about $0.0006 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 2026-09-03.