AgentSystemLabsOfficial·37 Agent Skills Included

core

Automated feature building, bug fixing, and code review pipelines

Turns plain-English coding goals into complete engineering workflows that classify intent, pick the right depth, and run the correct checks. Eliminates guessing which command to use, skipping security reviews, and shipping untested changes. Runs bundled playbooks and read-only reviewers for security, performance, and accessibility, then stops at verified code so you control publishing.

121|2

↗

npx skills add AgentSystemLabs/core --all -g -y

All Skills in This Repository (37)

Pure Emerald Level Indicators

📦 In Repo

AgentSystemLabs

check-release-risk

Analyze branch diffs to categorize release-breaking changes by risk.

commit

Split a dirty working tree into grouped, buildable git commits with pre-flight checks.

add-e2e-test

Add Playwright end-to-end tests with smoke gates and role-based assertions.

resolve-conflict

Resolve Git merge, rebase, or cherry-pick conflicts by inspecting marker intent and regenerating affected artifacts.

fix-pr-tests

Diagnoses GitHub CI test failures and applies minimal fixes to resolve them.

add-empty-error-states

Add empty and error UI states to TanStack Query and Router data-fetch surfaces.

audit-analytics

Audit analytics instrumentation for event naming consistency and PII leaks.

add-migration

Classify schema changes into additive, mutating, or destructive migration plans.

audit-seo-meta

Audit per-route SEO and social-share metadata across TanStack Start, Next.js, and HTML pages.

audit-perf

Audit route entry points for static performance bottleneck patterns.

audit-authz

Audits server-side authorization and ownership across handlers including tRPC and GraphQL.

add-regression-test

Generate regression tests that fail pre-fix and pass post-fix.

Official

Advanced

Frequently Asked Questions

FAQPage Schema

How to install AgentSystem?▼

Run `npx skills add AgentSystemLabs/core --all -g -y` in your terminal to install the full skill set globally.

What does the /ship command do?▼

It reads your plain-English goal, classifies it as create, fix, polish, remove, or audit, picks the right depth, and runs the matching engineering pipeline with built-in review gates.

Does AgentSystem commit or push my code?▼

No. It stops at a locally verified code candidate and hands the diff back to you, so committing, pushing, and opening pull requests stay under your control.

Which coding agents work with AgentSystem?▼

It works with Claude Code, Cursor, Codex, and OpenCode, since everything follows the standard SKILL.md format.

Do I need to memorize commands to use it?▼

No. You only describe your goal in plain English, and the single entry skill routes it to the right workflow and reviewers automatically.