core-review

core-review is a command for Claude Code from sanmak/specops. It costs 0 tokens per session (3,339 once invoked), scanned A, original, MIT.

A code-review command that checks changes against patterns specific to the SpecOps project, such as generated files, cross-platform behavior, and consistent variable use.

In plain words
What is it for?
Use it before merging code to inspect a pull request or local changes for known failure patterns.
Why use it?
It catches recurring project-specific mistakes that a general code-quality review may miss. It can review a pull request, staged changes, or all uncommitted changes.

Command for Claude Code

Part of the specops plugin — 2 skills, 14 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/sanmak/specops/core-review
Clone the repo
git clone --depth 1 https://github.com/sanmak/specops

Made for: Claude Code.

Or install specops, the plugin that ships this one along with the rest of its 2 skills, 14 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 core-review

README.md
[![agentmods](https://agentmods.dev/badge/commands/sanmak/specops/core-review.svg)](https://agentmods.dev/commands/sanmak/specops/core-review)
Your own site
<a href="https://agentmods.dev/commands/sanmak/specops/core-review"><img src="https://agentmods.dev/badge/commands/sanmak/specops/core-review.svg" alt="Measured on agentmods" height="20"></a>
Per session 0 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 3,339 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 $0.00000 $0.03339
Opus 5 $0.00000 $0.01670
Sonnet 5 $0.00000 $0.00668
Haiku 4.5 $0.00000 $0.00334

Measured 3d ago against content hash a6d46b5ed754, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

core-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.

.claude/commands/core-review.md · 307 lines

How it starts

The opening of the file, as written. The whole thing — 307 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Review code changes against SpecOps project-specific patterns. Catches recurring failure modes from real PRs — tool abstraction violations, generated file drift, cross-platform gaps, variable inconsistencies, and more. Complements full-review-gate (generic quality) and pr-fix (applying bot comments).

Instructions

Follow these steps precisely.

Argument Parsing

Parse $ARGUMENTS:

  • A number (e.g., 11) → PR mode: review that PR's diff
  • stagedStaged mode: review only staged changes (git diff --cached)
  • current or empty → Current mode: review all uncommitted changes (git diff HEAD)

Step 1: Collect the diff

PR mode:

  1. Verify gh --version is available. If not, report error and stop.
  2. Fetch the PR diff: gh pr diff <PR_NUMBER>
  3. Get changed file paths: gh pr view <PR_NUMBER> --json files -q '[.files[].path]'
  4. Save as DIFF_CONTENT and CHANGED_FILES.
  5. Set REVIEW_TARGET = "PR #<PR_NUMBER>"

Staged mode:

  1. git diff --cachedDIFF_CONTENT
  2. git diff --cached --name-onlyCHANGED_FILES
  3. Set REVIEW_TARGET = "staged changes"

Current mode:

  1. git diff HEADDIFF_CONTENT
  2. git diff HEAD --name-onlyCHANGED_FILES
  3. Set REVIEW_TARGET = "current changes"

If DIFF_CONTENT is empty, report "No changes found for REVIEW_TARGET" and stop.


Step 2: Run review passes

Run all four passes. Collect findings as a list with: [check_id], severity (P0/P1/P2/P3), file path, line (if known), description, and suggested fix.


Pass 1 — P0 Blockers

These indicate a broken build or a push that will definitely fail. Flag every instance.

[1a] Raw abstract operations in generated outputs

Read each generated platform file that appears in CHANGED_FILES OR read them fresh if Pass 1 is scanning for latent issues:

  • platforms/claude/SKILL.md
  • platforms/cursor/specops.mdc
  • platforms/codex/SKILL.md
  • platforms/copilot/specops.instructions.md
  • skills/specops/SKILL.md

Read the full file on GitHub · 307 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. 3d ago First seen · 307 lines · 0 tokens per session scan A a6d46b5ed754

Subscribe to this mod's changes

core-review is a command published in the GitHub repository sanmak/specops (49 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 3,339 tokens. 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.