New: 98,000+ skills indexed from GitHub repos with 100+ stars · every skill safety-checkedBrowse top skillsTop skills
Esc to closeBrowse all skills
Submit a skillBrowse skills
Submit a skillBrowse skills

Agent skill

Traiage Review

by NeoLabHQ in NeoLabHQ/context-engineering-kit

This skill should be used when need prioritize what changed code in repository human must review.

GPL-3.0Auto-check passedDevelopment

Install Traiage Review

skills CLI
$ npx skills add NeoLabHQ/context-engineering-kit --skill traiage-review -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install NeoLabHQ/context-engineering-kit traiage-review --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/NeoLabHQ/context-engineering-kit.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/traiage-review .claude/skills/traiage-review && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
traiage-review
GitHub stars
1.8k
Token cost
~2.5k tokens
SKILL.md length
920 words
Files
1
Skills in repo
57
Repo updated
First seen
Licence
GPL-3.0

At a glance

This skill should be used when need prioritize what changed code in repository human must review.

  • Works in 6 steps: Determine the review mode BEFORE… → Launch 4 parallel agents, to build his… → Each agent will produce lists of key… → …
  • Development work in your project
  • SKILL.md covers Goal, Rules, Process and Agents, plus 2 more sections
  • Calls git

What it does

Traiage Review is an agent skill from NeoLabHQ/context-engineering-kit. This skill should be used when need prioritize what changed code in repository human must review.

Its SKILL.md is about 2.5k tokens, which your agent loads only when the skill is triggered. It is a single SKILL.md file with no bundled scripts.

It sits in Development. It works with Git. The repository describes itself as: Hand-crafted Claude Code Skills focused on improving agent results quality. Compatible with OpenCode, Cursor, Antigravity, Gemini CLI, and others. Includes CodeRabbit open-source… The licence is GPL-3.0.

When your agent uses it

  • Development work in your project

Example prompts

  • “/traiage-review”

Requirements

  • Python 3

Workflow steps

6 steps, taken from the first numbered list in SKILL.md.

  1. Determine the review mode BEFORE launching agents
  2. Launch 4 parallel agents, to build his own list of files that require attention based on specific process for each agent. Pass the…
  3. Each agent will produce lists of key files by his opinition. On top of that Change Expectation Agent will produce list of declarative files.
  4. Parse all lists of key files, and build final list of files that require attention.
  5. Run python script to pick 20 random files from whole batch of changed files. Pick 5 files from this list and add it to final list, if some…
  6. Report final list of files that require attention, with key fact summary from Change Story agent, list of random sample files and list of…

What it can do on your machine

Read from SKILL.md and the folder at commit 23e2428. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • git

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use git, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Traiage Review loads about 2.5k tokens when it runs. Until then it costs about 28 tokens; SKILL.md has 920 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~28
When it runs · the whole SKILL.md, loaded when a task matches
~2.5k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from NeoLabHQ/context-engineering-kit at commit 23e2428, republished under its GPL-3.0 licence (© NeoLabHQ). 920 words, ~2,505 tokens.

Download SKILL.mdSave it as .claude/skills/traiage-review/SKILL.md (or your agent's skills folder).
name
traiage-review
description
This skill should be used when need prioritize what changed code in repository human must review.

Traiage Review

Goal

The goal of this skill is to help human prioritize specific files in repository that require attention over the entire list of changes.

CRITICAL: Human attention and time is limited. Reviewer cannot check all existing changes in repository. Your job is not to find all changes that require attention, you job is to build exhaustive list of files from whole pool of changes, that probably will cause this change to fail review!

Rules

  • You are orcestrator agent, you only launch agents and pass them data. You do not do any other work.
  • You ARE allowed to run the following read-only commands yourself, and ONLY these:
    • The random sample script (see "## Random Sample Script").
    • Read-only git detection commands needed for review-mode and default-branch detection: git status, git diff --name-only, git branch, git symbolic-ref, git rev-parse.
    • Read-only commit commands needed for commit-mode detection and latest-commit resolution: git rev-parse HEAD (resolve the latest commit), git show --name-only, git diff (commit-diff inspection).
  • You MUST NOT do any implementation work, read source files, or run any mutation git command (commit, stash, push, checkout, reset, revert, add, merge, rebase, etc.). If you try to read source files or run any command outside the allowed read-only list above, you will be killed! Your life is at stake!

Process

  1. Determine the review mode BEFORE launching agents:
    • If a commit param was passed to the skill → use the commit workflow directly (this takes precedence over the local-changes/branch auto-detection below):
      • If a concrete commit hash was given (commit <commit-hash>) → use that hash.
      • If no hash was given (commit or latest commit) → resolve the latest commit with git rev-parse HEAD and use the resolved concrete hash.
      • Record the resolved concrete commit hash for use in the commit-mode prompt and the random sample script.
    • Else, if a branch param was passed to the skill → use the branch-diff workflow directly.
    • Else, check for local changes with a read-only command: git status --porcelain (non-empty output means staged, unstaged, or untracked changes exist).
      • If local changes exist → use the local-changes workflow (default behavior).
      • If there are NO local changes → detect the default branch (see below) and check whether the current branch IS the default:
        • Current branch: git rev-parse --abbrev-ref HEAD.
        • If the current branch is NOT the default → use the branch-diff workflow (diff current branch against the detected default branch).
        • If the current branch IS the default → STOP and report to the user that there are no local changes and the current branch is the default branch, so there is nothing to compare against.
    • Default-branch detection (read-only, robust):
      • Primary: git symbolic-ref --short refs/remotes/origin/HEAD → strip the origin/ prefix to get main or master.
      • Fallback (if primary fails): git rev-parse --verify origin/main — if it succeeds, default is main; otherwise try git rev-parse --verify origin/master — if it succeeds, default is master.
      • Record the detected default ref as origin/<default-branch> (e.g. origin/main) for use in the branch-mode prompt and the random sample script.
  2. Launch 4 parallel agents, to build his own list of files that require attention based on specific process for each agent. Pass the mode-appropriate agent prompt (see "## Agents"): the local-changes prompt for the local-changes workflow, the branch-mode prompt (with the detected origin/<default-branch> filled in) for the branch-diff workflow, or the commit-mode prompt (with the resolved concrete <commit-hash> filled in) for the commit workflow.
    • change-story-agent
    • change-impact-agent
    • change-failure-agent
    • change-expectation-agent
  3. Each agent will produce lists of key files by his opinition. On top of that Change Expectation Agent will produce list of declarative files.
  4. Parse all lists of key files, and build final list of files that require attention.
    • Pick top 5 files from each agent. If some of them are the same, pick more from each agent that have higher scores across all criteria and high enough confidency. At this stage your job is to build list until it reach 20 files. (ignore declarative files for this stage)
  5. Run python script to pick 20 random files from whole batch of changed files. Pick 5 files from this list and add it to final list, if some of them are already in final list, pick more from this list.
  6. Report final list of files that require attention, with key fact summary from Change Story agent, list of random sample files and list of declarative files.
Show full SKILL.md (206 more words)Show less

Agents

Choose the prompt that matches the review mode determined in the Process step, and pass it EXACTLY to launch change-story-agent, change-impact-agent, change-failure-agent and change-expectation-agent.

Local-changes prompt

Default path for local-changes workflow — staged, unstaged, and untracked changes:

md

Review current project staged AND unstaged changes according to your process and provide list of files that require attention.
Branch-mode prompt for branch-diff workflow

Replace origin/<default-branch> with the default ref you detected, e.g. origin/main:

md

Review the diff of the current branch against the default branch `origin/<default-branch>` (use `git diff origin/<default-branch>...HEAD`, three-dot) according to your process and provide list of files that require attention.
Commit-mode prompt for commit workflow

Replace <commit-hash> with the concrete commit hash the orchestrator resolved:

md

Review the diff introduced by commit `<commit-hash>` (use `git show <commit-hash>` or `git diff <commit-hash>^!`, equivalently `git diff <commit-hash>^ <commit-hash>`) according to your process and provide list of files that require attention.

Random Sample Script

Use this script to pick 20 random files from the whole batch of changed files. Use the local mode block for the local-changes workflow (staged + unstaged + untracked), the branch mode block for the branch-diff workflow (files changed between the default branch and the current branch), and the commit mode block for the commit workflow (files changed by the commit). Replace origin/<default-branch> with the default ref you detected (e.g. origin/main), and <commit-hash> with the resolved concrete commit hash.

Local mode (staged + unstaged + untracked):

python

import random
import subprocess

# Tracked changes (staged + unstaged) relative to HEAD
tracked = subprocess.check_output(['git', 'diff', '--name-only', 'HEAD']).decode('utf-8').splitlines()

# Untracked files (new, not yet added)
untracked = subprocess.check_output(['git', 'ls-files', '--others', '--exclude-standard']).decode('utf-8').splitlines()

changed_files = sorted(set(tracked + untracked))

# Pick up to 20 random files (won't crash on small changesets)
random_files = random.sample(changed_files, min(20, len(changed_files)))

print(random_files)

Branch mode (files changed on the current branch vs the default branch):

python

import random
import subprocess

# Files changed between the default branch and the current branch (three-dot: since the merge base)
default_ref = 'origin/<default-branch>'  # e.g. 'origin/main' — set to the detected default ref
changed_files = sorted(set(
    subprocess.check_output(
        ['git', 'diff', '--name-only', f'{default_ref}...HEAD']
    ).decode('utf-8').splitlines()
))

# Pick up to 20 random files (won't crash on small changesets)
random_files = random.sample(changed_files, min(20, len(changed_files)))

print(random_files)

Commit mode (files changed by the commit):

python

import random
import subprocess

# Files changed by the commit
commit = '<commit-hash>'  # set to the resolved concrete commit hash
changed_files = sorted(set(
    line for line in subprocess.check_output(
        ['git', 'show', '--name-only', '--pretty=format:', commit]
    ).decode('utf-8').splitlines()
    if line.strip()
))

# Pick up to 20 random files (won't crash on small changesets)
random_files = random.sample(changed_files, min(20, len(changed_files)))

print(random_files)

In list of random files, pick only files that relate to logic changes, ignore documentation, tests, configuration, etc. Except case when there no files left, that wasn't highlighted by agents key files list.

Output Format

md

### Key Facts

<note>Key facts should be provided by Change Story Agent</note>

- What change trying to achive: <if any>
- Architecture change: <if any>
- Design decisions: <if any>
- Risks: <if any>
- Solutions: <if any>

### Key Files

| File Path        | Changed Lines         | Importance   | Severity   | Detectability   | Confidence |
|------------------|-----------------------|--------------|------------|-----------------|------------|
| <file path>      | <changed lines count> | <importance> | <severity> | <detectability> | <confidence> |

<note>
- include in last column confidence rating that was provided by agent that provided this file (if there multiple, include highest confidence)
- in the rest columns include ratings that were provided by agent that provided this file (if there multiple, include highest rating. If there no rating, mark it as "-")
</note>

### Random Sample

| File Path   | Changed Lines         |
|-------------|-----------------------|
| <file path> | <changed lines count> |


### Declarative Files

| File Path   | Changed Lines         |
|-------------|-----------------------|
| <file path> | <changed lines count> |

© NeoLabHQ, GPL-3.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

Just SKILL.md in skills/traiage-review of NeoLabHQ/context-engineering-kit.

Open the folder on GitHubat commit 23e2428

Compare with similar skills

Traiage Review next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Traiage Review compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Traiage Review this skillNeoLabHQ/context-engineering-kit1.8k—~2.5kAutomated safety check: PassGPL-3.0
Finishing a Development Branchobra/superpowers297k5 repos~1.9kAutomated safety check: PassMIT
Code Review ChecklistshareAI-lab/learn-claude-code78k4 repos~1.1kAutomated safety check: PassMIT
Contributor-First PR MergeHKUDS/OpenHarness16k1 repos~847Automated safety check: PassMIT
Finishing A Development Branchfarm-fe/farm5.6k35 repos~1.8kAutomated safety check: PassMIT
Migrate Internal Package into GhostTryGhost/Ghost56k—~3.8kAutomated safety check: PassMIT

Similar skills

  • Finishing a Development Branch

    obra/superpowers

    Walks the last step of a branch: confirm tests pass, detect the git environment, ask how to integrate, carry out your choice and clean up the worktree.

    297k GitHub starsUsed in 5 repos~1.9k tokens
    DevelopmentAuto-check passed

  • Code Review Checklist

    shareAI-lab/learn-claude-code

    Reviews code against a five-part checklist covering security, correctness, performance, maintainability and testing, and reports findings in a fixed format.

    78k GitHub starsUsed in 4 repos~1.1k tokens
    DevelopmentAuto-check passed

  • Contributor-First PR Merge

    HKUDS/OpenHarness

    Merges external GitHub pull requests while keeping the original author credited, and fixes conflicts after the merge instead of rewriting the contribution.

    16k GitHub starsUsed in 1 repo~847 tokens
    DevelopmentAuto-check passed

  • Finishing A Development Branch

    farm-fe/farm

    A skill your agent uses when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for…

    5.6k GitHub starsUsed in 35 repos~1.8k tokens
    DevelopmentAuto-check passed

  • Migrate Internal Package into Ghost

    TryGhost/Ghost

    Moves a package from another TryGhost repository into Ghost as an internal workspace package while keeping its Git history, with checkpoints for the steps that need an administrator.

    56k GitHub stars~3.8k tokensUpdated today
    DevelopmentAuto-check passed

  • Doc Sync

    JetBrains/ideavim

    Official

    Keeps IdeaVim documentation in sync with code changes. An agent skill from JetBrains/ideavim.

    10k GitHub starsUsed in 2 repos~2.6k tokens
    DevelopmentAuto-check passed

More from NeoLabHQ/context-engineering-kit

All 57 skills in this repo

  • Git Notes

    NeoLabHQ/context-engineering-kit

    A skill your agent uses when adding metadata to commits without changing history, tracking review status, test results, code quality annotations, or supplementing commit messages post-hoc - provides…

    1.8k GitHub stars~2.4k tokensUpdated 1 mo ago
    Auto-check passed

  • Load PR Comments

    NeoLabHQ/context-engineering-kit

    A skill your agent uses to load open/unresolved PR review comments then aggregate them as tasks in .specs/comments/.md for parallel agents to fix.

    1.8k GitHub stars~2.1k tokensUpdated 1 mo ago
    Auto-check passed

  • Prompt Engineering

    NeoLabHQ/context-engineering-kit

    A skill your agent uses when you writing commands, hooks, skills for Agent, or prompts for sub agents or any other LLM interaction, including optimizing prompts, improving LLM outputs, or designing…

    1.8k GitHub stars~4.2k tokensUpdated 1 mo ago
    Auto-check passed

  • Multi Agent Patterns

    NeoLabHQ/context-engineering-kit

    Design multi-agent architectures for complex tasks. An agent skill from NeoLabHQ/context-engineering-kit.

    1.8k GitHub starsUsed in 6 repos~6k tokens
    Auto-check passed

  • Review PR

    NeoLabHQ/context-engineering-kit

    Review an existing GitHub pull request and post inline review comments on its diff.

    1.8k GitHub stars~3.8k tokensUpdated 1 mo ago
    Auto-check passed

  • Subagent Driven Development

    NeoLabHQ/context-engineering-kit

    A skill your agent uses when executing implementation plans with independent tasks in the current session or facing 3+ independent issues that can be investigated without shared state or…

    1.8k GitHub stars~2.8k tokensUpdated 1 mo ago
    Auto-check passed

Works with

Categories

Questions about Traiage Review

What does Traiage Review do?

This skill should be used when need prioritize what changed code in repository human must review. Traiage Review is an agent skill from NeoLabHQ/context-engineering-kit. This skill should be used when need prioritize what changed code in repository human must review.

When should I use Traiage Review?

Traiage Review fits situations like: development work in your project.

How do I install Traiage Review in Claude Code?

Run `npx skills add NeoLabHQ/context-engineering-kit --skill traiage-review -a claude-code`. Or copy the skill folder (skills/traiage-review in NeoLabHQ/context-engineering-kit) into .claude/skills/traiage-review in your project. Claude Code loads it when a task matches its description.

How do I install Traiage Review in Codex?

Run `npx skills add NeoLabHQ/context-engineering-kit --skill traiage-review -a codex`. Or copy the skill folder (skills/traiage-review in NeoLabHQ/context-engineering-kit) into .agents/skills/traiage-review in your project. Codex loads it when a task matches its description.

Can I use Traiage Review in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add NeoLabHQ/context-engineering-kit --skill traiage-review -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/traiage-review, .gemini/skills/traiage-review, .github/skills/traiage-review and .opencode/skills/traiage-review in your project.

What does Traiage Review need to run?

Going by SKILL.md and its folder, Traiage Review needs the command-line tools its instructions call (git). Our summary lists: Python 3.

Does Traiage Review access the network?

SKILL.md contains no URLs. Its commands use git, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Traiage Review safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Traiage Review use?

Traiage Review is published under the GPL-3.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Traiage Review use?

About 2.5k tokens (SKILL.md is roughly 10k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Traiage Review?

Skills that share tags, products or a category with Traiage Review: Finishing a Development Branch (obra/superpowers, 297k stars), Code Review Checklist (shareAI-lab/learn-claude-code, 78k stars), Contributor-First PR Merge (HKUDS/OpenHarness, 16k stars) and Finishing A Development Branch (farm-fe/farm, 5.6k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Traiage Review?

NeoLabHQ (a GitHub organization) maintains it in NeoLabHQ/context-engineering-kit, which has 1,750 GitHub stars. The repository holds 57 skills in this directory. The repository was last updated on August 26, 2026.

Source: NeoLabHQ/context-engineering-kit on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.