Teams need repeatable ways to package workflows and domain knowledge for AI agents. This skill guides Claude Code through designing, writing, and testing focused Agent Skills.

Agent systems can fail through silent routing, transport, or configuration issues. This skill guides structured checks for processes, logs, environment, and message delivery.

Teams lose context when AI agents use local todo lists and ad hoc Git commands. This skill directs agents to Ship CLI tasks, stacked changes, PR creation, and guarded webhook follow-up.

NixOS and Home Manager module setup can be repetitive and error-prone. This skill gives concise patterns for packages, modules, and imports.

Frontend bugs can be hard to reproduce from code alone. This skill guides Claude, Codex, and Claude Code through Chrome DevTools MCP evidence gathering.

Blog teams often lose consistency when each post uses different fields, tone, and SEO details. This skill guides Claude, Codex, or Claude Code to draft posts that match the existing Next.js BlogPost pattern.

New ZenTao users often struggle to connect roles, objects, and command-line actions. This skill guides a role-based tour through real ZenTao workflows.

ZenTao data is difficult to query consistently across projects, bugs, tasks, requirements, and tests. This skill maps user requests to zentao CLI commands and response formats.

Teams need fast access to ZenTao work items without manually building API requests. This skill maps natural language requests to ZenTao REST API v2.0 calls.

Agents can miss required skill workflows when tasks start quickly. This skill gives a clear process for finding, announcing, and following relevant skills.

Process skills often fail when agents face time pressure, authority pressure, or sunk cost. This skill gives a repeatable subagent test workflow to expose failures and improve instructions.

Poor tests often verify mocks instead of real application behavior. This skill gives Claude, Codex, and Claude Code a checklist for safer TDD, focused mocks, and cleaner test design.

Reusable skills often stay local because contributors lack a clear publishing workflow. This skill gives Claude, Codex, and Claude Code a structured process for preparing a branch, commit, push, and pull request.

Complex bugs often appear far from their original trigger. This skill guides backward tracing, targeted instrumentation, and test isolation to find the source.

Chrome extension teams need safe Manifest V3 architecture, permissions, and review support. This skill guides Claude, Codex, and Claude Code through secure design, implementation, testing, and publishing tasks.

Sales teams often lose time researching accounts that do not match their market. This skill builds focused lead lists with fit scores and outreach guidance.

Low-quality screenshots can make documentation, presentations, and social posts hard to read. This skill guides enhancement, upscaling, sharpening, artifact reduction, and output choices.

Invalid data often bypasses a single validation point and fails later in execution. This skill guides Claude, Codex, and Claude Code to add layered checks that make recurring bugs harder to reintroduce.

Long-form writing often stalls when research, structure, citations, and revision happen in separate steps. This skill guides Claude, Codex, or Claude Code through outlining, source gathering, drafting feedback, and final polish.

Flaky tests often fail because they guess how long async work takes. This skill shows Claude, Codex, and Claude Code how to wait for real conditions instead of fixed delays.