review — for Claude Code review, community, for Claude Code, ide skills, strict: true, is always, Architecture, CLAUDE_SKILL_DIR, Version, Audits

v1.0.0

Über diesen Skill

Geeigneter Einsatz: Ideal for AI agents that need echo "architecture docs: ${claude skill dir}/architecture/". Lokalisierte Zusammenfassung: Run your own background coding agents on Kubernetes - isolated pods, injected credentials, scheduled triggers, Slack-native. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

Funktionen

echo "Architecture docs: ${CLAUDE SKILL DIR}/architecture/"
echo "Version: $(cat "${CLAUDE SKILL DIR}/VERSION")"
Phase 1 — Get Version
Phase 2 — Generate Checklist
You (the main agent) generate a strict checklist from the architecture docs.

# Core Topics

kagenti kagenti
[8]
[1]
Updated: 4/24/2026

Killer-Skills Review

Decision support comes first. Repository text comes second.

Reference-Only Page Review Score: 10/11

This page remains useful for teams, but Killer-Skills treats it as reference material instead of a primary organic landing page.

Original recommendation layer Concrete use-case guidance Explicit limitations and caution Quality floor passed for review
Review Score
10/11
Quality Score
70
Canonical Locale
en
Detected Body Locale
en

Geeigneter Einsatz: Ideal for AI agents that need echo "architecture docs: ${claude skill dir}/architecture/". Lokalisierte Zusammenfassung: Run your own background coding agents on Kubernetes - isolated pods, injected credentials, scheduled triggers, Slack-native. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

Warum diese Fähigkeit verwenden

Empfehlung: review helps agents echo "architecture docs: ${claude skill dir}/architecture/". Run your own background coding agents on Kubernetes - isolated pods, injected credentials, scheduled triggers, Slack-native.

Am besten geeignet für

Geeigneter Einsatz: Ideal for AI agents that need echo "architecture docs: ${claude skill dir}/architecture/".

Handlungsfähige Anwendungsfälle for review

Anwendungsfall: Applying echo "Architecture docs: ${CLAUDE SKILL DIR}/architecture/"
Anwendungsfall: Applying echo "Version: $(cat "${CLAUDE SKILL DIR}/VERSION")"
Anwendungsfall: Applying Phase 1 — Get Version

! Sicherheit & Einschränkungen

  • Einschraenkung: Each item must be verifiable by reading files (checking imports, config values, directory structure, etc.)
  • Einschraenkung: Launch a subagent (using the Agent tool) to perform the actual audit. The subagent prompt must:
  • Einschraenkung: Checklist rules ONLY include rules explicitly stated in the architecture docs

Why this page is reference-only

  • - Current locale does not satisfy the locale-governance contract.

Source Boundary

The section below is imported from the upstream repository and should be treated as secondary evidence. Use the Killer-Skills review above as the primary layer for fit, risk, and installation decisions.

After The Review

Decide The Next Action Before You Keep Reading Repository Material

Killer-Skills should not stop at opening repository instructions. It should help you decide whether to install this skill, when to cross-check against trusted collections, and when to move into workflow rollout.

Labs Demo

Browser Sandbox Environment

⚡️ Ready to unleash?

Experience this Agent in a zero-setup browser environment powered by WebContainers. No installation required.

Boot Container Sandbox

FAQ & Installation Steps

These questions and steps mirror the structured data on this page for better search understanding.

? Frequently Asked Questions

What is review?

Geeigneter Einsatz: Ideal for AI agents that need echo "architecture docs: ${claude skill dir}/architecture/". Lokalisierte Zusammenfassung: Run your own background coding agents on Kubernetes - isolated pods, injected credentials, scheduled triggers, Slack-native. This AI agent skill supports Claude Code, Cursor, and Windsurf workflows.

How do I install review?

Run the command: npx killer-skills add kagenti/humr. It works with Cursor, Windsurf, VS Code, Claude Code, and 19+ other IDEs.

What are the use cases for review?

Key use cases include: Anwendungsfall: Applying echo "Architecture docs: ${CLAUDE SKILL DIR}/architecture/", Anwendungsfall: Applying echo "Version: $(cat "${CLAUDE SKILL DIR}/VERSION")", Anwendungsfall: Applying Phase 1 — Get Version.

Which IDEs are compatible with review?

This skill is compatible with Cursor, Windsurf, VS Code, Trae, Claude Code, OpenClaw, Aider, Codex, OpenCode, Goose, Cline, Roo Code, Kiro, Augment Code, Continue, GitHub Copilot, Sourcegraph Cody, and Amazon Q Developer. Use the Killer-Skills CLI for universal one-command installation.

Are there any limitations for review?

Einschraenkung: Each item must be verifiable by reading files (checking imports, config values, directory structure, etc.). Einschraenkung: Launch a subagent (using the Agent tool) to perform the actual audit. The subagent prompt must:. Einschraenkung: Checklist rules ONLY include rules explicitly stated in the architecture docs.

How To Install

  1. 1. Open your terminal

    Open the terminal or command line in your project directory.

  2. 2. Run the install command

    Run: npx killer-skills add kagenti/humr. The CLI will automatically detect your IDE or AI agent and configure the skill.

  3. 3. Start using the skill

    The skill is now active. Your AI agent can use review immediately in the current project.

! Reference-Only Mode

This page remains useful for installation and reference, but Killer-Skills no longer treats it as a primary indexable landing page. Read the review above before relying on the upstream repository instructions.

Upstream Repository Material

The section below is imported from the upstream repository and should be treated as secondary evidence. Use the Killer-Skills review above as the primary layer for fit, risk, and installation decisions.

Upstream Source

review

Run your own background coding agents on Kubernetes - isolated pods, injected credentials, scheduled triggers, Slack-native. This AI agent skill supports

SKILL.md
Readonly
Upstream Repository Material
The section below is imported from the upstream repository and should be treated as secondary evidence. Use the Killer-Skills review above as the primary layer for fit, risk, and installation decisions.
Supporting Evidence
echo "Architecture docs: ${CLAUDE_SKILL_DIR}/architecture/"
echo "Version: $(cat "${CLAUDE_SKILL_DIR}/VERSION")"

TSEng Review

Audits an existing TypeScript project against the architecture rules using a two-phase approach to prevent hallucination. Each review produces an immutable record appended to tseng/reviews/.

Phase 1 — Get Version

The current tseng version is shown above (injected from the VERSION file). This version is embedded in the review record so every checklist is traceable to the architecture revision that produced it.

Phase 2 — Generate Checklist

You (the main agent) generate a strict checklist from the architecture docs.

  1. Read architecture/index.md from the architecture docs directory (shown above).
  2. Read every file linked from the index (in the same architecture docs directory).
  3. Extract every concrete, verifiable rule from the docs. Each rule becomes a checklist item.

Checklist rules

  • ONLY include rules explicitly stated in the architecture docs. Do not infer, extrapolate, or add "best practices" that aren't written down.
  • Do not include tseng/ files (index.md, project-structure.md, adoption.md, reviews/) as checklist items — those are generated outputs, not architecture rules.
  • Each item must be verifiable by reading files (checking imports, config values, directory structure, etc.)
  • Use the exact terminology from the docs.
  • Group items by source file (Stack, Architecture, Project Structure).

Phase 3 — Assign Review Number & Write Record

Determine the next review number:

  1. Check if tseng/reviews/index.md exists in the target project. If it does, read it to find the highest existing review number.
  2. If no index exists, the next number is 001.
  3. Otherwise increment: if the last review is 003, the next is 004.

Write the checklist to tseng/reviews/NNN.md using this format:

markdown
1# Review #NNN 2 3<!-- tseng_version: {version} --> 4<!-- status: open --> 5<!-- created: {YYYY-MM-DD} --> 6 7Generated from architecture docs (tseng v{version}). 8 9## Stack 10- [ ] Uses tRPC for API layer 11- [ ] Uses Zod for input validation 12- [ ] TypeScript strict mode enabled (`strict: true` in tsconfig) 13- ... 14 15## Architecture 16- [ ] ... 17 18## Project Structure 19- [ ] ...

The status is always open for review — the review skill is read-only and does not lock records. Only adopt and upgrade lock records.

Phase 4 — Update Reviews Index

Write or update tseng/reviews/index.md:

markdown
1# Review History 2 3| # | Date | TSEng Version | Status | 4|---|------|---------------|--------| 5| [001](001.md) | 2026-04-12 | 1.0.0 | open |

Append the new row. Never modify existing rows.

Phase 5 — Audit via Subagent

Launch a subagent (using the Agent tool) to perform the actual audit. The subagent prompt must:

  • Include the full contents of the generated checklist (copy-paste it into the prompt)
  • Tell the subagent the project root path
  • Instruct the subagent to go through each checklist item, check pass/fail, and cite specific files/lines as evidence
  • Do NOT include architecture docs, skill context, or any other instructions in the subagent prompt

Use this exact subagent prompt template:

You are auditing a TypeScript project at {project_path} against a checklist of architecture rules.

Go through EVERY item below. For each one:
- Check the project files to determine if the rule is satisfied
- Mark it ✅ (pass) or ❌ (fail) or ⚠️ (partially met / not applicable)
- Cite the specific file and line that proves your assessment
- If it fails, briefly say what's wrong

Do NOT check anything beyond this list. Do NOT add suggestions, improvements, or opinions. Only verify what's listed.

CHECKLIST:
{checklist_contents}

Phase 6 — Report

Take the subagent's output and present the final report to the user:

  • Summary — one-line overall assessment
  • Passes — rules the project satisfies
  • Violations — rules that are broken, with file/line citations from the subagent
  • Suggestions — actionable fix for each violation

Do not update the review record with audit results — the review skill is read-only. The record stays open with unchecked boxes. Only adopt and upgrade mark items and lock records.

Guidelines

  • This is a read-only audit by default. Don't modify files unless the user asks you to fix violations.
  • The server runtime is pluggable (Hono or Express). Accept either as valid.
  • Package names are flexible. Do not flag packages for having non-standard names — audit their internal structure and dependencies instead. A project may have multiple server and client packages.
  • If a tseng/project-structure.md exists, check whether it is outdated and offer to update it.
  • Do not flag the absence of tseng/ — it is a generated folder.
  • If the project doesn't use the expected stack at all, say so and suggest whether bootstrapping from scratch would be more appropriate.
  • Review records are immutable once written. Never modify a previously written review file. Always append a new record.

Verwandte Fähigkeiten

Looking for an alternative to review or another community skill for your workflow? Explore these related open-source skills.

Alle anzeigen

openclaw-release-maintainer

Logo of openclaw
openclaw

Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞

333.8k
0
Künstliche Intelligenz

widget-generator

Logo of f
f

Erzeugen Sie anpassbare Widget-Plugins für das Prompts.Chat-Feed-System

149.6k
0
Künstliche Intelligenz

flags

Logo of vercel
vercel

Das React-Framework

138.4k
0
Browser

pr-review

Logo of pytorch
pytorch

Tensor und dynamische neuronale Netze in Python mit starker GPU-Beschleunigung

98.6k
0
Entwickler