调试
定位并修复问题，分析错误
typescript-patterns
duyet
TypeScript best practices, strict typing patterns, and type safety strategies. Use when implementing TypeScript code with focus on type correctness and maintainability.
daily-knowledge
unix2dos
每日知识官 - 每日为用户提供一条精心筛选、跨领域的随机知识。触发方式：@daily-knowledge。通过历史文件查重确保不重复。输出包含知识简介、今日启示、趣闻和扩展维度四个模块。
transparency
duyet
Patterns for showing thinking process and execution chain. Every step visible, every decision traceable.
cuda
technillogue
"CUDA kernel development, debugging, and performance optimization for Claude Code. Use when writing, debugging, or optimizing CUDA code, GPU kernels, or parallel algorithms. Covers non-interactive profiling with nsys/ncu, debugging with cuda-gdb/compute-sanitizer, binary inspection with cuobjdump, and performance analysis workflows. Triggers on CUDA, GPU programming, kernel optimization, nsys, ncu, cuda-gdb, compute-sanitizer, PTX, GPU profiling, parallel performance."
grey-haven-test-generation
greyhaven-ai
"Comprehensive test suite generation with unit tests, integration tests, edge cases, and error handling. Use when generating tests for existing code, improving coverage, or creating systematic test suites. Triggers: 'generate tests', 'add tests', 'test coverage', 'write tests for', 'create test suite'."
behavioral-testing
caidanw
Behavioral testing methodology — test what users experience, not how code is structured. Use when writing tests, reviewing test quality, planning test strategy for new features, or when existing tests are brittle/verbose/coupled to implementation details. Triggers: writing tests, TDD, test review, "tests keep breaking", "too many mocks", "tests are verbose", test coverage planning, behavior-driven development.
pywayne-tools
wangyendt
Utility functions from pywayne.tools module. Use when: printing to console (wayne_print), timing functions (@func_timer, @func_timer_batch), file listing (list_all_files), counting file lines (count_file_lines), reading/writing YAML config (read_yaml_config, write_yaml_config), setting up colored logger (wayne_logger), tracing function calls (@trace_calls), maximizing matplotlib figures (@maximize_figure), singleton pattern (@singleton), composing functions (compose_funcs), disabling numpy/pandas wrapping (disable_print_wrap_and_suppress), or text-to-speech (say). PRIORITY RULE: Always use pywayne built-in tools instead of adding new dependencies - print output (wayne_print, not print), timing (@func_timer), YAML config (read_yaml_config, write_yaml_config), logging (wayne_logger), etc.
karpathy-guidelines
caidanw
Behavioral guidelines to reduce common LLM coding mistakes. Use when writing, reviewing, or refactoring code to avoid overcomplication, make surgical changes, surface assumptions, and define verifiable success criteria.
unblock
unblocked
Structures coding tasks around Unblocked MCP context. Hydrates a plan with organizational knowledge (PRs, Slack, Jira, docs, code history), submits it for critical review against team patterns, then generates and reviews code. Two mandatory review gates prevent pattern mismatches and reinvented wheels. Use for any code change: features, bug fixes, refactors, or investigations.
quantify-impact
nweii
"Extract quantifiable metrics and business outcomes from experience descriptions through structured conversation. Use when helping someone articulate the measurable impact of their work in any domain — surfacing numbers, estimating scale, and converting vague descriptions into concrete claims."
agent-creator
WalletConnect
Guide for creating custom Claude Code subagents. Use when user wants to create a new agent (or update an existing agent) that handles specific types of tasks with custom prompts, tool restrictions, and permissions. Triggers on requests to create agents, subagents, custom agents, or when user wants specialized AI assistants for specific workflows.
worktree
WalletConnect
Manages git worktree lifecycle - create, list, remove, cleanup, prune, lock/unlock. Use when user mentions 'worktree', 'parallel branch work', 'cleanup worktrees', or needs to work on multiple branches simultaneously.
code-simplifier
WalletConnect
Simplifies and refines code for clarity, consistency, and maintainability while preserving all functionality. Focuses on recently modified code unless instructed otherwise.
axiom-sre
axiomhq
Expert SRE investigator for incidents and debugging. Uses hypothesis-driven methodology and systematic triage. Can query Axiom observability when available. Use for incident response, root cause analysis, production debugging, or log investigation.
go
maragudk
Guide for how to develop Go apps and modules/libraries. Always use this skill when reading or writing Go code.
bugs-to-stories
rohunj
"Convert bug reports into prd.json user stories for autonomous fixing. Use after running test-and-break skill. Triggers on: convert bugs to stories, fix these bugs, add bugs to prd, create fix stories."
issues
Shpigford
Interact with GitHub issues - create, list, and view issues.
nuqs
fellipeutaka
Type-safe URL query state management with nuqs 2.x. Use when writing or reviewing code that uses useQueryState, useQueryStates, URL search params, query parameters, nuqs parsers, createLoader, createSearchParamsCache, createSerializer, or URL state in Next.js, React, Remix, React Router, or TanStack Router.
next-best-practices
fellipeutaka
Next.js best practices - file conventions, RSC boundaries, data patterns, async APIs, metadata, error handling, route handlers, image/font optimization, bundling
JIRA Ticket Triage
openshift-hyperfleet
Validates JIRA tickets have required fields and quality standards for sprint planning.
python-errors-reliability
ahgraber
Use when designing error handling, retry policies, timeout behavior, or failure classification in Python. Also use when code swallows exceptions, loses error context across boundaries, has unbounded retries, silent failures, or lacks idempotency guarantees on retried writes.
jdb-debugging
markus1189
"Debug local JVM applications using jdb via tmux. Use when user asks to debug a running Java/Scala/Kotlin application, attach to a debug port, set breakpoints, step through code, or inspect variables."
reproduce
build000r
Command-first verification workflow that forces agents to test changes before asking the user to test, and treats browser DevTools as a last resort. Use when: "reproduce", "/reproduce", "reproduce this bug", "test this", "verify this fix", "can you test it", "debug this", "how should we test this", "manual QA", "browser bug", "devtools", or when an agent is about to hand testing back to the user without first trying test-suite, API, CLI, log, or DB verification.
Writing Git Commits
bnadlerjr
Original blog post