注释
代码注释与标注
go-api-documentation
sentenz
Automates API documentation creation for Go projects using godoc conventions and best practices. Use when creating, modifying, or reviewing documentation comments, or when the user mentions godoc, documentation, or API docs.
qq-napcat-adapter
eamars
用 NapCat / OneBot v11 正向 WebSocket 接通宿主 QQ 通道（私聊 + 四个授权群）：入站事件先落盘再进宿主，公开出站只来自宿主 outbox，发送结果按平台真实响应回报执（unknown 不重发），发送后还会用 get_msg 把自己那条读回来核对段序列（仅观测）；会读清对方是谁（昵称、当前群名片、群身份）并且改名换名片后仍认得是同一个人；私聊出站可带一张宿主指定的图（字节只从宿主通道 API 取，群图未开）；含授权闸门（explicit 显式名单与 automatic 自动准入分开）、群成员快照与 at/reply 规范化、共用账号约束与日志读法；离线自检跑在内置占位夹具上（9000000xx 占位号），不依赖本机私有配置文件。
code-review-master
data219
Perform senior-level code and pull request reviews with explicit checks for architecture and design fit, security and performance risks, code quality, and functional correctness. Use when asked to review a PR, commit diff, patch, or code snippet and provide a structured decision with actionable findings.
triage
bakabala27-svg
Move issues and external PRs through a state machine of triage roles, categorise, verify, grill if needed, and write agent-ready briefs. Use this to categorize or evaluate an incoming issue.
review
brandonocasey
"Review PRs/MRs, diffs, commits, branches, or threads for verified defects; supports fixes and review loops."
engineering-standards-rust
pragmatic-engineer
Rust rules for code and reviews, covering ownership, errors, types, API design, async, unsafe, testing and Clippy lints. Use when writing or reviewing Rust, or a diff that touches .rs files.
Commenting Intent
obra
Comment WHY code exists and non-obvious decisions, not WHAT code does (mechanics)
Writing Evergreen Comments
obra
Write comments explaining WHAT and WHY, never temporal context or history
grounding-review
pragmatic-engineer
Evidence and labeling rules for reviewing a PR or diff. Use for any quick or deep code review. Not for investigation.
writing-style
pragmatic-engineer
"One plain, simple-English standard for prose a human reads: chat, PR descriptions, commit messages, review comments, tickets, Confluence, Slack, ADRs. Cuts filler, binary contrasts and dashes. Use before writing or posting any."
engineering-standards-python
pragmatic-engineer
Python rules for code and reviews, covering style, idioms, types, errors, async, testing, packaging, security and the Ruff rules that check them. Use when writing or reviewing Python, or a diff that touches .py or .pyi files.
engineering-standards
pragmatic-engineer
Team standards for code design, commits, PRs, testing, and deployment. Use when designing code, preparing a PR, or planning tests or a deploy.
refine
weside-ai
Writes a build-ready story plan from front-loaded context, with no user in the room — what a dispatched refiner runs. Triggers: "/we:refine", "refine this story without asking me".
develop
weside-ai
Dev-only worker: implements its chunk in its own worktree, runs local gates, commits, reports, stops — no PR, no CI, no ticket; pushes only on the brief's word. Triggers: "/we:develop", "implement only", "dev worker", or an /we:orchestrate dispatch.
epic
weside-ai
Epic (Solo) at the Initiative altitude: status from plan + ticketing mirror, or refine/create. Triggers: "/we:epic", "epic", "refine epic", "new epic".
ci-review
weside-ai
Collects every CI and PR-review finding, fixes by severity, resolves bot threads, pushes once. Triggers: "/we:ci-review", "fix ci", "fix reviews", "ci failed".
hiring-helper
dazuck
Screen job candidates using AI-powered evaluation. Use when asked to "screen candidates", "review applicants", "evaluate candidates", or when reviewing resumes and applications.
sdlc-review
magnusthovsland
Review Kanban handoffs and route verified outcomes.
technical-orientation
dazuck
Explain technical projects, codebases, tools, or documentation to a non-engineer. Use when the user says "orient me", "explain this repo", "what does this do", "get me up to speed", "break this down for me", or shares a GitHub URL asking for explanation. Also use when a non-technical person needs to understand something technical quickly.
check-reviews
kbrdn1
"Analyser et appliquer les reviews IA — en ligne d'abord (@codex review en commentaire PR / Codex Cloud, prioritaire, charge locale nulle), puis CLI locaux (Codex, CodeRabbit) et bots GitHub (Copilot/CodeRabbit) en fallback — et maintainers sur une PR"
analyze-pr
SiinXu
Analyze a GitHub pull request for necessity, description completeness, verification evidence, risks, and merge readiness. Pairs with review-pr for the blocking-vs-nit checklist.
close
DmitryDF
Close session - create diary entry, archive logs, git commit
codebase-visualizer
kbrdn1
Transforme un dépôt en carte d'architecture isométrique interactive (HTML autonome, thème Claude Dark), publiée en artifact — blocs dimensionnés aux vraies LOC, arêtes dérivées du graphe d'imports mesuré, points de données animés, drill-down "go inside", trace pas-à-pas, mode diff entre deux révisions, plus un dossier technique mesuré (couches, cycles de dépendances, hubs, manifestes, chaîne d'outils, CI, binaires appelés). Déclencheurs — "visualise ce codebase", "carte isométrique du repo", "atlas du codebase", "diagramme visuel de l'architecture", "codebase atlas", /me:codebase-visualizer. Sert aussi à mettre à jour ou étendre une carte existante. Pas pour un schéma de mécanisme isolé dans une réponse en prose (SVG inline), pas pour du Mermaid, pas pour une maquette d'UI produit.
premortem
kbrdn1
"Exécute un premortem sur n'importe quel plan, lancement, produit, recrutement, stratégie ou décision. Suppose que ça a déjà échoué 6 mois plus tard et remonte dans le temps pour trouver tous les motifs. Produit un plan révisé avec les angles morts exposés. DÉCLENCHEURS OBLIGATOIRES : 'premortem ça', 'premortem mon', 'exécute un premortem', 'qu'est-ce qui pourrait tuer ça', 'stress-teste ce plan', 'qu'est-ce que je rate ici', 'trouve les angles morts'. DÉCLENCHEURS FORTS : 'qu'est-ce qui pourrait mal tourner', 'est-ce que je rate quelque chose', 'troue ce plan', 'où ça va casser', 'avocat du diable'. NE PAS déclencher sur des demandes simples de retour, des questions factuelles ou des demandes au Conseil LLM. DÉCLENCHER quand quelqu'un a un plan ou un engagement où le coût de se tromper est élevé."