LLM & Agents

Formats git commit messages following Conventional Commits and respects user's attribution preferences from CLAUDE.md

Prompt engineering framework for Claude Code. Transforms vague requests into precise TCRO-structured prompts. Use when: crafting prompts for code generation, improving prompts not getting results, structuring multi-step workflows, debugging pattern drift. Triggers: optimize prompt, improve prompt, create prompt, prompt engineering, TCRO, prompt template, better prompt.

Structured plan-writing skill adapted from obra/superpowers. Produces actionable plans that map directly to claude-ctx workflows and tasks.

Break LLM name defaults with external entropy. Use when character names cluster around statistical medians (Chen, Patel, Maya, Marcus), when cast has collision risks, or when fantasy cultures need phonologically consistent naming.

Isolated git worktree workflow for coding agents. Handles worktree creation, atomic commits, branch publishing, optional PR creation, and cleanup. Use when implementing code changes that need branch isolation. Trigger terms - worktree, isolated workspace, branch isolation, atomic commits, code changes in isolation, parallel agent work.

Build comprehensive randomization lists for creative entropy. Use when you need to create or expand lists of story elements (professions, locations, objects, names, etc.) for use with entropy tools. Leverages research sources like Kiwix/Wikipedia to build lists with good variety and size.

Applies disciplined test-driven development (TDD) using Red-Green-Refactor cycles with modern TypeScript/Bun and Rust tooling. Use when implementing features with TDD, writing tests first, refactoring with test coverage, bug fixes requiring test reproduction, or when TDD, test-first, red-green-refactor, testing methodology, or --tdd flag are mentioned. Covers test quality, mutation testing, and both TypeScript and Rust workflows.

Configure and troubleshoot TypeScript testing with NX and Vite. Use this skill when working with test configuration, coverage reporting, or debugging test execution issues in TypeScript projects using NX monorepo structure with Vite/Vitest test runner.

Use when generating tests for new or existing code to improve coverage - provides a structured workflow for analyzing code, creating tests, and validating coverage goals.

This skill should be used when the model needs to ensure code quality through comprehensive linting and formatting. It provides automatic linting workflows for orchestrators (format → lint → resolve via concurrent agents) and sub-agents (lint touched files before task completion). Prevents claiming "production ready" code without verification. Includes linting rules knowledge base for ruff, mypy, and bandit, plus the linting-root-cause-resolver agent for systematic issue resolution.

Prompt engineering standards and context engineering principles for AI agents based on Anthropic best practices. Covers clarity, structure, progressive discovery, and optimization for signal-to-noise ratio.

Systematic pre-commit quality gate using checklist-based review methodology with announcement protocol. Use when reviewing code before commit, performing quality gates, conducting systematic code review, or when fresh-eyes, pre-commit-review, quality-gate, or --review are mentioned. Reviewer agent should load this for thorough reviews.

Guide for creating effective skills. This skill should be used when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.

Record Claude Code sessions with asciinema. TRIGGERS - record session, asciinema record, capture terminal, record claude, demo recording, record ASCII, ASCII terminal, terminal screen capture, shell screen capture, ASCII screen capture, screen recording. Generates ready-to-copy commands with dynamic workspace-based filenames.

Database schema (users, messages, photos tables), environment variables, ImageKit storage structure, authentication flow, and service integrations. Use when working on database operations, API logic, authentication, or infrastructure.

Centralized colored output formatter for all skills, agents, and commands with ANSI escape codes

Provides structured error handling protocol for all agent operations. Ensures users receive clear error reports, troubleshooting guidance, and actionable next steps. Use when any tool operation fails, encounters an error, or produces unexpected results.

Patterns for delegating UI work to specialized agents. Covers synthesis-master vs specialized agents, multi-agent UI generation workflows, and orchestration strategies for complex UI tasks.

全范围召回 - 从 Project CLAUDE.md、User CLAUDE.md 和 Memory MCP 查询历史知识

Guide for creating effective skills. Use when users want to create a new skill (or update an existing skill) that extends Claude's capabilities with specialized knowledge, workflows, or tool integrations.