Latest Agent Skills
Everything added to the registry recently, newest first.
- jeffersonvalente SkillHelp
Reads the project state and suggests the next Cardume command or skill for the request at hand. Use when you don't know what to run next, or automatic routing picked nothing.
- jeffersonvalente SkillGate
Checks every acceptance criterion against executed evidence and writes a report in docs/gates/ with an APPROVE/REJECT recommendation. Use when a level-3 phase or milestone ends or the user asks if something is ready to release.
- jeffersonvalente SkillPrd
Writes a PRD in docs/prd/ with metrics, acceptance criteria, no-gos, risks and blockers. Use for level-3 work: a new system or product phase, or when the user asks for a PRD; feature-level work uses spec.
- jeffersonvalente SkillAdr
Records a decision in docs/adr/ with options and consequences, and updates the index. Use when a technical decision is made or changed, even in a small task: database, framework, integration, data model or a notable trade-off.
- matmar10 SkillFolder Labels
Print folder labels in two sizes, each with its own calculation and default dimensions. "manila" is a stick-on tab label on the adhesive roll, size 3 down to 2 text, upright or sideways, padded to the label's length. "hanging" is a paper insert for a hanging folder's plastic tab, on a non-adhesive roll, size 2 only, kept to the insert's width even on the full 80mm roll. Either can be sized to the user's own label dimensions. Also handles a whole filing system in batches, one roll at a time, resumable after a stock-out. TRIGGER when the user asks to label or print labels for folders, file folders, manila/manilla folders, hanging folders, hanging files, suspension files, Pendaflex tabs, tab inserts, folder tabs or a filing system, or gives a label size (e.g. "50 x 20 mm", "2 by 3/4 inch") for tab text.
- matmar10 SkillGrocery List
Turn recipes or a menu into a shopping list grouped by grocery store department (produce, bakery, deli, butcher, seafood, dairy, canned goods, dry goods, baking, spices, frozen...) in walking order, with duplicate ingredients merged, then print it on the thermal receipt printer. Takes a recipe or menu URL, pasted recipe text, or several of them at once. TRIGGER when the user asks for a shopping or grocery list from a recipe, menu, meal plan or link, asks "what do I need to buy for...", or wants a recipe's ingredients printed.
- matmar10 SkillThermal
Print on the user's ESC/POS thermal receipt or label printer — shopping and to-do lists, agendas, jar/bin/box labels, folder tabs, notes, tickets, barcodes, QR codes and images. Compose a job from primitives (text sizes, big bitmap text, rules, rows, space, barcode, QR, image, cut), preview it against the paper, then print. Also sets the printer up the first time (find it, pick a model, measure the paper), records which roll is loaded, checks for paper, and saves reusable print templates. TRIGGER when the user asks to print something, make or print a label or sticker, print a list or receipt, label a jar/bin/folder, or mentions a thermal, receipt or label printer.
- mneira10 BundleUpcodes Pr Review
Review an UpCodes PR by checking out its latest version and fanning out one subagent per check (readability, simplicity and footguns, reuse of existing patterns, style guide, bugs, and an independent security analysis, plus tests, migrations, and AI checks when relevant), then merge their findings into actionable local comments. On explicit request, adds the comments to the user's pending (unsubmitted) GitHub review for them to edit and submit.
- minqiyang BundleOpen In Rex
Use when the user wants to continue a Claude desktop app session in the Rex terminal (the CLI): 在 Rex 里打开这个会话, 在 CLI 里续上, 在终端里继续这个会话, 把这个会话转到 Rex, 复制续接命令, open this session in Rex, continue this in the CLI. Opens a new Rex tab that runs claude --resume for this session, or for another session the user names by its sidebar title. Can also copy the resume command.
- thatsheaps BundleAjay Prakash Playbook
Ajay Prakash's operator playbook, distilled from his public newsletter (Ajay's Quest / Lab Notes), ajayslist.com and his LinkedIn posts. Its core is his async comms framework: flagposting plus the six labelled update types (FYI, Approval, Decision, Action, CTL, Feedback) and the what / why / so-what frame. It also covers his methods for meetings, hiring and managing teams, growth and copywriting, ideas and strategy docs, and building with AI agents. Use it whenever someone wants to write or rewrite a Slack, Teams or email update, a weekly or 1:1 update to a manager, a decision request, a status report, a complaint, or a meeting agenda, even if they never mention Ajay. Also use it when they ask how to cut meetings, write a strategy doc, hire, scale or manage a team, diagnose a conversion or growth problem, write landing page or ad copy, or set up an AI agent team, or when they mention "flagposting", "CTL", "closing the loop", "Ajay's Quest", "Ajay API" or Ajay Prakash.
- codepilots SkillWardrobe Skill Integration
Use when adding or wiring a new M21-style agentic wardrobe skill (app/skills/*).
- codepilots BundleMeta Instagram Integration
Use when integrating Instagram/Meta data or DMs.
- codepilots SkillMeta Hackathon Compliance
Use when working on meta-styles' Meta hackathon entry.
- codepilots SkillDemo Launch
Use when launching, restarting, or verifying the meta-styles local demo stack (mock, live, or mock-remote).
- codepilots BundleMuse Code
Use when delegating a task to Meta's Muse Code CLI.
- edgardoacosta BundleCbq Concurrency Test
Concurrent live test (real HTTP calls, NOT Pest) of every command in manufacturing-core-pkg that uses the EventCommands/CBQ toolkit — inventory movements over Kafka, idempotency, redelivery, and the Redis commands. Use when the user asks to "test all the commands", "run the concurrency test", "check the engines still work" after changes to the toolkit or the engines.
- edgardoacosta SkillLegacy Pr UI Review
Review and test changes that port a legacy screen into manufacturing-core-pkg — either a PR (ours or a teammate's) or local code not pushed yet (a bug branch, uncommitted work). Legacy is the source of truth, every claim is validated against legacy code or the DB, the definitive test is a Playwright run on the real screen, and the output is a pass/fail matrix. Use when the user asks to "test this PR", "review Julio's/Jorge's changes", "test my changes", "is the fix done?", "run the full test", or "run the Obsidian bug list against this screen".
- edgardoacosta SkillBoui Adminlte Migration Conventions
Conventions for migrating legacy AdminLTE admin screens into BOUI/Uxmal. Use when a developer needs practical guidance to reinterpret AdminLTE pages using Uxmal Design System 2026, BOUI components, Laravel page structure, and clear backoffice UX patterns without requiring pixel-perfect validation or specialized tooling.
- edgardoacosta BundleManufacturing
EnviaFlores Manufacturing migration context and playbooks. Use for anything about the manufacturing team or its GitHub project 59 - creating or filling Epic/Story/Task/Bug issues, researching legacy screens (talleres, suministros, OPS, admin-api), Grafana/Loki/Tempo telemetry for ef-manufacturing or the app backend, checking what each dev (Jorge, Julio, Guillermo, Dali) is working on or whether they have tasks, assigning QA, reading the Notion migration databases or the Obsidian manufacturing notes, reviewing/testing teammates' PRs, shipping (commit/PR, core-pkg release, host composer bump, permissions), and the workshop print servers (workshop-daemon, labeler, Zebra/CUPS queues, ZPL label templates, etiquetas QR, print server deploys and "users can't print" diagnosis).
- edgardoacosta BundleBump Core Pkg
Bump manufacturing-core-pkg in the ef-manufacturing host to a released tag and push main (staging deploy). Use when Edgardo says "new version of core vX.Y.Z", "bump core", "deploy core to staging". Looks up the latest tags, confirms the tag with him, regenerates composer.json/lock from HEAD, commits "Bump manufacturing-core-pkg to vX.Y.Z" with no co-author and pushes main only. Never tags or releases the host.
- kvothesson SkillKvothesson
Sistema operativo del universo creativo Kvothesson. Activar SIEMPRE que el usuario mencione: "Kvothesson", "El Gran Presidente", "cyberpunk argentino", "folklore digital", "Pombero", "Cerebro Austral", "Argentina post-singularidad", "Rabbi Borgeson", "Bucle Cerati", "Hombres-Tanque", "Ruta 404", "Tribus 2030", "Kill Switch Criollo", "Gaucho Cuántico", "crear episodio", "escribir descripción de video", "expandir lore", "nuevo capítulo de la serie", "guión para el canal", "título para el video", "comentario del canal", "universo kvothesson", "Mythos", "Protocolo Glasswing", "La Gran Digitalización", "ciclos", "Dark-outs", "Glitchers", "Constructor de Horizontes", "Auri", "Rendous", "Aegis", "Vesper", "Nyx", "Aleph", "mechas neuronales", "laboratorio flotante", "firma de mecha", "arquitecto del silicio", o cualquier combinación de folklore argentino + tecnología/IA/cyberpunk. También activar cuando el usuario pide escribir contenido para YouTube en tono cyberpunk, quiere desarrollar el lore de presidentes argent
- fnvasque SkillPublicar
Use when Felipe escribe «/publicar <video>», «publica este video», «programa este reel» o entrega un video vertical ya listo para subirlo a @ia.punto.es (revisa el video, pregunta cuándo, escribe el caption, muestra todo, espera el OK y lo programa).
- fnvasque BundleCaso
Use when Felipe escribe «/caso», «cuenta cómo hice…», «arma el caso de este proyecto» o pide convertir un proyecto hecho con Claude en una historia paso a paso (PDF, carrusel y reel sin voz, con la música del proyecto si la hay) para @ia.punto.es.
- aia-studio BundleTech Book Writer
用 agent 流水线把技术主题写成一本结构化的书(Markdown 书稿 + 可运行代码 + 图表源码)。当用户要写技术书、写手册、写系列教程、把一个主题组织成 8 章以上的体系化长文,或已有零散内容需要成书时使用。关键词:写书、技术书、手册、系列教程、电子书、book、handbook、成书。
- will12193 SkillMaintain Authoring Agent Instructions And Skills
William's preferences for how agent instruction files and skills are structured, sized and scoped: one lean always-loaded instruction file that orchestrates scoped, self-maintaining skills. Use when: about to create or edit any agent file (`*.instructions.md`, `SKILL.md`, `AGENTS.md`, `copilot-instructions.md`), in any repo or globally, including adding a single entry or gotcha; deciding whether content belongs in the instruction file or a skill; splitting, trimming or restructuring agent files; the instruction file is getting long; setting up agent files for a repo that has none; reviewing agent files against these preferences. This is a policy layer: it does not cover file-format mechanics (frontmatter syntax, applyTo globs, debugging why a file isn't loaded), which belong to any built-in customization tooling. Distilled from the platform-api setup.
- will12193 SkillMaintain Global Agent Config
Rules for evolving William's GLOBAL agent files (instructions/ and skills/, sourced from the ai hub repo and mirrored into ~/.copilot): when to add or change a rule or skill, what scope belongs there, and the mandatory pull, disclosure and push steps. Use when: the user says 'you should always...', 'never...', 'from now on', corrects a wrong assumption or preference that applies to all repos, a knowledge gap or recurring pattern is discovered that is relevant to all coding regardless of repo, or you are about to create/edit/delete any file under ~/.copilot/instructions or ~/.copilot/skills, regardless of which tool or skill makes the edit. This is a policy layer (when/whether to change, disclosure, sync); it does not cover file-format mechanics, which belong to any built-in customization tooling.
- will12193 SkillMaintain AI Hub Backup Repo
Maintain the user's AI hub repo (GitHub will12193/ai), the source of truth for the global skills and instructions that are mirrored into ~/.copilot on every machine. Use when: the user asks to add, change, rename or remove any Copilot skill, instruction, or MCP config; is about to edit anything under ~/.copilot/skills or ~/.copilot/instructions (pull first, edit in the repo); mentions the 'ai hub', 'ai repo', 'sync.sh', 'sync.ps1', or backing up or syncing Copilot config between machines.
- will12193 SkillAlign Repo Agent Files
Bring the current repo's instruction files and skills in line with William's authoring conventions without losing existing knowledge. Use when: the user says 'align this repo', 'restructure this repo's agent files', 'bring the repo's skills/instructions up to my conventions', or asks to tidy, split or standardise a repo's existing instruction and skill files.
- will12193 SkillDiff Code Review V2
Exhaustive v2 review of the code changes on the current branch (git diff against a base branch, default origin/master): reports every confirmed issue with a severity tag, reviewing file by file. Use when: the user explicitly asks for 'diff code review v2', 'diff-code-review-v2' or 'exhaustive diff review'. For a plain 'review the branch/diff', use `diff-code-review` instead.
- will12193 SkillGitlab Mr Summary
Generate a GitLab merge request title, summary and manual test plan from the git diff of the current branch against a base branch (default origin/master), then offer to update the branch's existing MR with it. Use when: the user asks for an MR description, MR summary, merge request text, PR description, a test plan for their branch, or to update/refresh the MR.
- will12193 SkillDiff Code Review
Review the code changes on the current branch (git diff against a base branch, default origin/master) for performance, bug risks, security, code quality, regressions and best practices. Use when: the user asks to review the branch, review the diff, review their changes before opening an MR, or says 'diff code review' / 'code review this branch'.
- will12193 SkillHealth Check
Report honestly what the agent currently has in its context: instruction files loaded (grouped by level: global, repo, other), skills available, and relevant session state. Use when: the user says 'health check', 'health-check', '/health-check', 'what do you have loaded', 'what instructions/skills do you have', 'what's in your context', or asks for a status of the session.
- will12193 SkillPlanning
Plan a piece of work as a markdown plan spec saved in the repo's tmp folder, then grill William on it with the `grilling` skill before any implementation. Complements the built-in planning tools rather than replacing them. Use when: the user mentions planning in any form ('plan this', 'make a plan', 'let's plan', 'plan mode', 'how would you approach', 'write a spec for').
- eric-lingren BundleImprove Codebase Architecture
Find deepening opportunities in a codebase, informed by domain vocabulary and ADR decisions loaded via the context-loader agent. Use when the user wants to improve architecture, find refactoring opportunities, consolidate tightly-coupled modules, or make a codebase more testable and AI-navigable.
- eric-lingren BundleImprove Directory Structure
Build a dependency graph over a directory tree and surface reorganization opportunities (misplaced shared nodes, low-cohesion clusters, near-duplicate files) informed by domain vocabulary and ADR decisions. Grills each candidate against a curated litmus set, then hands off a verified reorg plan via to-seed. Never moves files or rewrites references directly. Use when the user wants to reorganize a directory, audit repo layout, find directory-structure smells, or asks "does this belong here".
- eric-lingren SkillImprove Agent Benchmarks
Cheap iterative improvement of a shared agent. Runs the next stage for the chosen agent (contract, cases, iterate). Contract stage drafts a cited, history-validated contract.json with an Opus subagent, shows each check with its source, historical pass rate and real failing examples, and freezes it on approval. Use when the user invokes /improve-agent-benchmarks [agent-name].