Depuració
Detecció i correcció de problemes i anàlisi d’errors
makepad-reference
ZhangHanDong
CRITICAL: Use for Makepad troubleshooting and reference. Triggers on: troubleshoot, error, debug, fix, problem, issue, no matching field, parse error, widget not found, UI not updating, code quality, refactor, responsive layout, adaptive, api docs, reference, documentation, 故障排除, 错误, 调试, 问题, 修复
calculator
Code-and-Sorts
Performs arbitrary-precision arithmetic calculations including addition, subtraction, multiplication, division, and exponents. Use when the user asks to calculate, compute, or evaluate math expressions, or when precise decimal arithmetic is needed to avoid floating-point errors.
release
colonelpanic8
Use when user asks to release, publish, bump version, or prepare a new version for deployment
gilfoyle
axiomhq
SRE agent that does what you can't. Queries your observability stack. Finds root causes. Doesn't panic. Doesn't guess. Doesn't care about your feelings. Use for incident response, debugging, root cause analysis, or log analysis.
go-best-practices
NeverSight
Provides Go patterns for type-first development with custom types, interfaces, functional options, and error handling. Must use when reading or writing Go files.
e2e
NeverSight
Run e2e tests, fix flake and outdated tests, identify bugs against spec. Use when running e2e tests, debugging test failures, or fixing flaky tests. Never changes source code logic or API without spec backing.
python-best-practices
NeverSight
Provides Python patterns for type-first development with dataclasses, discriminated unions, NewType, and Protocol. Must use when reading or writing Python files.
git-best-practices
NeverSight
Git workflow patterns for commits, branching, PRs, and history management across heterogeneous repositories. Use when creating commits, managing branches, opening pull requests, or rewriting history. Do not use for non-git implementation tasks or repo-specific release policy decisions without repository documentation.
op-cli
NeverSight
Secure 1Password CLI patterns for reading secrets, discovering vaults/items, and piping credentials to other tools. Use when reading from 1Password, rotating secrets, or piping credentials to wrangler/kubectl/etc. Triggers on op CLI, 1Password, secret rotation, or credential piping tasks.
testing-best-practices
NeverSight
Test layering, execution, and CI guidance across unit, integration, and e2e. Use when designing tests, writing test cases, or planning test strategy for a module.
yellow-best-practices
NeverSight
Yellow Network and Nitrolite (ERC-7824) development best practices for building state channel applications. Use when building apps with Yellow SDK, implementing state channels, connecting to ClearNodes, managing off-chain transactions, or working with application sessions.
code-review-thorough
p-wegner
In-depth AI code review using a more capable model — catches subtle bugs and architecture issues
code-review
p-wegner
Default AI code review prompt — customize per project to change review behavior
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.
download-file
accomplish-ai
'Download files in Chrome on Windows and macOS. Handles triggering downloads, detecting and resolving popups, verifying completion, and locating files.'
git-commit
accomplish-ai
Create well-structured git commits with conventional commit messages, proper staging, and commit best practices.
diagnose
griffinwork40
"Parallel root-cause analysis for bugs and failing tests. Use when a test fails, a bug is reported, or behavior is unexplained — dispatches sub-agents to form and validate hypotheses in isolated worktrees."
ai-evals
oldwinter
"Create an AI Evals Pack (eval PRD, test set, rubric, judge plan, results + iteration loop). Use for LLM evaluation, benchmarks, rubrics, error analysis/open coding, and ship/no-ship quality gates for AI features."
go-expert
nguyenthienthanh
"Go/Golang backend expert. PROACTIVELY use when working with Go, Gin, Echo, Fiber frameworks. Triggers: golang, go, gin, echo, fiber"
react-expert
nguyenthienthanh
"React best practices expert. PROACTIVELY use when working with React components, hooks, state management. Triggers: React, JSX, hooks, useState, useEffect, component"
review-changes
wcygan
Orchestrate multiple review agents for a comprehensive pre-PR code review. Analyzes diff, verifies intent alignment with linked issues, and checks for broader codebase impacts. Use before opening a PR to catch bugs, security issues, silent failures, and performance problems. Keywords: review, code review, pre-PR, diff review, check changes, audit changes
async-rust
wcygan
Async Rust and Tokio ecosystem expert. Provides idiomatic patterns, architectural guidance, and code review for async Rust code using tokio, futures, bytes, tracing, tonic, hyper, tower, axum, tokio-stream, and pin-project. Covers runtime internals, cancellation safety, graceful shutdown, channel selection, shared state, backpressure, structured concurrency, error handling, Tower middleware, stream processing, and performance tuning. Also identifies async footguns (blocking the runtime, holding locks across .await, silent task panics, accidental cancellation) and suggests fixes. Use when writing async Rust, debugging async issues, reviewing async code, choosing between spawn/join/select, designing channel topologies, implementing graceful shutdown, or learning how futures, Pin, and the Tokio runtime work under the hood. Keywords: async, await, tokio, futures, spawn, select, join, channels, streams, pin, tower, tonic, hyper, axum, tracing, bytes, cancellation, runtime, backpressure, graceful shutdown.
launch-ready
wcygan
Vague recommendations: "Monitor closely" is not a mitigation plan
fix-issue
wcygan
Take a GitHub issue number, investigate it, implement a fix, write tests, and open a PR. End-to-end workflow from assigned issue to merged PR. Use when you want to fix a specific GitHub issue. Keywords: fix issue, github issue, resolve issue, close issue, implement issue, issue to PR, bug fix