Productivity

syp1xd Skill

Verify a lead is a real, direct buyer before any outreach - check the company website, who owns the requirement, and any mandatory vendor qualifications. Use before writing or sending a cold email to a new prospect.

ryangalea23 Skill

Capture and find notes. Three types: meeting (notes from a specific meeting), idea (something to think about later), fact (a tidbit to remember). All notes save to a notes folder (default `~/work/notes/`, see Storage below) with required frontmatter and consistent filenames so search actually works. Search runs across the whole notes root (notes, and anything else you keep under that same folder, like saved emails or briefs). Use when asked to "take a note", "note this", "save this idea", "save this fact", "meeting notes for...", "find my notes about...", "what did I write down about...", "show meeting notes", "show ideas", "show facts", or invoked as /notes. Also surfaces for natural phrases like "I should write this down" or "remind me of this".

kakatur Bundle

Polish existing written prose or spoken narration so it sounds natural while preserving facts, technical meaning, structure, links, code, and author voice. Use when asked to humanize or polish a draft, or when another selected workflow requires a final language pass. Do not use for initial drafting, fact-checking, authorship disguise, or substantive content invention.

kristenpan Bundle

The single complete Pickoo workflow for new Xiaohongshu, WeChat, Douyin, Bilibili, Xiaoyuzhou, podcast, music, article/blog, or general web links. Use for 收藏、保存、收录、抓取、归档、拆解、分析、OCR、ASR or 完整处理 from Codex, WorkBuddy, Hermes, Claude, or Claudian. It owns intake and the complete downstream workflow; obsolete standalone capture or link-intake skills must not be selected.

kristenpan Bundle

Process, validate, repair, or resume an existing Pickoo source package after save-to-favorites or the shared task worker hands it off. Handles evidence-preserving capture, OCR, ASR, chaptering, analysis, and quality-gated assets inside that package. Do not use this skill as the intake route for an ordinary new link; route new 收藏、保存、收录、抓取、归档 or 完整处理 requests through save-to-favorites.

acevinh Skill

Small code task where the user ALREADY NAMED ≤3 files: bug fix, UI tweak, add a button/column/filter, small refactor. Scan only what they gave you, implement, verify. Triggers on fix, chỉnh, sửa, thêm, bỏ, update, 'small change', 'chỉnh lại', 'sửa giúp', 'thêm cho tôi', 'bỏ cái này', 'update lại', a pasted error with a file path, or 1-3 file paths plus a short instruction. Boundary: no files named, >3 files, spans multiple packages/repos, or unknown root cause → escalate to a full scan-and-plan workflow instead. File count decides, not how small it feels.

timothystewart6 Bundle

Take the current local Git branch from finished implementation through verification, commit, push, pull request creation, CI repair, code review, final rebase, conversation resolution, and squash merge. Use when the user asks to finish, ship, PR, review, validate, or merge work that already exists in a local Git repository. The workflow verifies the actual diff, runs the repository's existing checks, diagnoses failures to root cause, adds regression tests for real bugs, optionally requests GitHub Copilot review exactly once, can perform an independent adversarial review, waits for CI, and merges only after all gates are clean.

mimmo-the-root Skill

General token-efficiency habits for this kit's agents — prefer targeted search over full reads, precise @-mentions, periodic summarization, avoid re-reading known context. Read once at session start or when a task looks like it'll involve many files/tool calls.

acevinh Bundle

DEFAULT router for real work: scan → classify → match existing patterns → plan with self-critique → TODO → implement → dispatch parallel slices → verify-debug loop. Use when the task touches >3 files, spans more than one repo or package, names no files, or the root cause is not yet known. Triggers on scan, todo, fix bug, implement feature, refactor, add tab/button/column, edit UI, 'similar to X', 'tương tự như', 'triển khai', 'update lại', 'chỉnh sửa', 'làm tiếp', or a design image / URL to test. Only ≤3 already-named files go to quick-fix. Answers in Vietnamese when the user writes Vietnamese.

isuzu-shiranui Bundle

Drive the open Unity Editor from the shell with the isuzu-unity-cli command: console errors, compiling and tests, scenes, GameObjects and components, prefabs, play mode, uGUI clicks, and C# run inside the Editor. SKILL.md gives a command for each common task, so no tool list is needed.

iabuelruzz Bundle

Use when working on a Python project with tests, configuration files, generated artifacts, GUI code, parser/filter logic, or reporting workflows. Provides a generic workflow for scoped edits and validation.

ankit1598 Skill

Before editing or referencing any file, discover its current path with Glob or Grep, then Read that path immediately before StrReplace or Write. Never use remembered paths or stale buffer from a prior turn; user may move files or edit on disk between prompts. Use at the start of every task that touches an existing file, including skills under ~/.agents/skills/.

bvlcq Skill

Cria uma ação A-#### a partir do que o dono digitou: '/todo preciso ler X até quarta' → ação no ledger com deadline resolvido, pessoas e projeto linkados. Use quando o dono disser 'preciso', 'tenho que', 'lembra de', ou pedir criação direta de tarefa.

idk-brasil Bundle

Use when the user wants a meeting recording turned into a speaker-labeled transcript and a structured "ata" (meeting minutes, PT-BR), or invokes /ata. Especially when they ask to transcribe/gerar ata of "the latest video" without naming a file.

arin117kaushik Bundle

Use when the user wants a weekly or monthly Substack content calendar, a plan for Notes and posts, a posting cadence, section-by-section planning for a multi-section publication, or asks to schedule or automate a run of Substack content. Not for writing the pieces themselves (use the writer skills).

sumanjeet0012 Skill

Unified Tailscale network operations router and master runbook. Guides secure device connectivity, Tailscale SSH execution, peer-to-peer file transfer (Taildrop), private service sharing (Tailscale Serve), and public internet publishing (Tailscale Funnel). Use when the user asks for general "tailscale help", "connect to tailscale", "tailscale status", "tailscale setup", or wants to perform any administrative task across their Tailnet.

agentcorp-inc Bundle

MUST activate when the project contains a uiBundles/*/src/ directory and scaffolding a new UI bundle or app, or when editing ui-bundle.json, .uibundle-meta.xml, or CSP trusted site files. Use this skill when scaffolding with sf template generate ui-bundle, configuring ui-bundle.json (routing, headers, outputDir), or registering CSP Trusted Sites. Activate when the task involves files matching *.uibundle-meta.xml, ui-bundle.json, or cspTrustedSites/*.cspTrustedSite-meta.xml.

rospocroccante Skill

Use when the user asks for meeting talking points, dev-meeting bullets, standup notes, or any brief they will read aloud (Italian or English), and when they push back that a draft sounds robotic, like an NPC, or like a changelog.

dangtran1003 Bundle

Từ 1 ticket ISD ($ISD_URL) viết trang KB trên $WIKI_URL, gắn link KB vào ticket rồi đẩy ticket qua Resolve → KB Update → W/A for KB → W/A FOR CLOSE. Dùng khi user nói "tạo KB cho ticket ISD-xxxxx", "viết KB rồi đóng ticket", "gắn KB vào task", "kéo ticket sang W/A for close". Trigger - /isd-kb, "tạo trang kb cho ISD-xxxxx".

franky47 Bundle

Turborepo monorepo build system guidance. Triggers on: turbo.json, task pipelines, dependsOn, caching, remote cache, the "turbo" CLI, --filter, --affected, CI optimization, environment variables, internal packages, monorepo structure/best practices, and boundaries. Use when user: configures tasks/workflows/pipelines, creates packages, sets up monorepo, shares code between apps, runs changed/affected packages, debugs cache, or has apps/packages directories.

bb-dev-aa Skill

Single session-long orchestrator for this workspace - loads your open Asana work (or a named task/project), resolves the client project (store handle + local folder) and starts/reuses the dev server, then - when run again once work looks done - drafts a commit message, pushes, opens a PR, and files DQA/FQA QA subtasks with the preview link. Replaces start-day, task, and shopify-project-resolution as separate skills; this is the only entry point for "what am I working on", "start my day", loading a specific Asana task's context, or resolving which project/store/folder is active. Session-scoped - it tracks the active task and resolved project from earlier in this same conversation, it does not persist across separate Claude Code sessions. Use when the user says "let's work", "start my day", references a task name/keyword/Asana URL, or re-invokes /lets-work partway through a task to move it to QA.

jakeschincariol Skill

Plan a week or a month of YouTube uploads - what to post, when, and in what order, sized to the creator's actual capacity. Use for "plan my week", "content calendar", "what should I post", "I have no idea what to make next".

umer-2612 Bundle

Professional Git and GitHub workflow conventions, including CI configuration. Use this skill for every version control or repository operation — staging changes, writing commit messages, creating or switching branches, opening pull requests, rewriting history, tagging releases, setting up branch protection, authoring or reviewing GitHub Actions workflows, or any `git` command. Apply these rules mechanically to keep `main` deployable, commits atomic and conventional, history navigable, PRs reviewable in under thirty minutes, and CI fast and reliable. Trigger this whenever you're about to commit, push, open or merge a PR, run `rebase`, `reset`, `cherry-pick`, write a `.github/workflows/*.yml` file, or configure repository settings — even for trivial one-line changes. Free-text prose in commit bodies and PR descriptions is run through the `umer-tone` skill before it ships.

franky47 Skill

Pick one open task and implement it end-to-end as a unit of work. Use when the user asks to pick a task.

shwsgithub Skill

백엔드 워크플로우

shwsgithub Skill

프론트 워크플로우

ntphat0322 Bundle

Always use this skill before claiming any work is done, fixed, passing, or complete — even if it seems obvious. Also use when receiving code review feedback from any source (human or automated), when finishing a task or major feature, before committing or creating a PR, or when about to express satisfaction with work. Covers three practices: receiving feedback with technical rigor (no performative agreement), triggering code review via /code-review command or cook/fix pipelines, and verification gates (run the command, read the output, then make the claim). Never skip — the cost of false completion claims is higher than the cost of verifying. References: code-review-reception.md, requesting-code-review.md, verification-before-completion.md.

ntphat0322 Bundle

Complete browser automation with Playwright. Auto-detects dev servers, writes clean test scripts to /tmp. Test pages, fill forms, take screenshots, check responsive design, validate UX, test login flows, check links, automate any browser task. Use when user wants to test websites, automate browser interactions, validate web functionality, or perform any browser-based testing.

ntphat0322 Skill

Suggests manual context compaction at logical intervals to preserve context through task phases rather than arbitrary auto-compaction.

specsnl Skill

Use this skill to convert raw client input (email, WhatsApp, verbal, or a Linear issue URL) into a properly structured Linear issue in Specs style. Provide the raw text or a Linear issue URL and this skill will draft a developer-ready ticket with IST/SOLL structure, acceptance criteria, and QA steps.

sayamsingla2000 Skill

Interview the user relentlessly about a plan or coding task until reaching shared understanding at the level of files, signatures, and edge cases, resolving each branch of the decision tree. Use when user wants to stress-test a plan, get grilled on their design, align before implementation, or mentions "grill me".

panmax Skill

Structured code-review workflow ported from earendil-works/pi-review (Codex-style rubric). Use when the user wants a code review — review uncommitted changes, review against a base branch, review a specific commit, review a GitHub pull request, or snapshot-review folders/files. Also handles the post-review handoff ("end-review" summary) and "fix review findings". Triggers: /pi-review, /skill:pi-review, /review, review uncommitted, review branch main, review commit abc123, review pr 123, review folder src docs, 代码评审, 评审一下, 帮我 review, end-review, 修复评审问题.

jahangirhussen Skill

Enforce test-driven development (write the test first, watch it fail, then make it pass) for production code, bug fixes, and refactors. Make sure to use this skill WHENEVER writing or changing production code, fixing a bug, or refactoring — even outside a formal spec-driven workflow, and even if the user doesn't say "test". It is reusable on its own. Do NOT force it on exploratory/throwaway/analysis code where the correct output isn't known in advance (there, use data-validation, regression/snapshot, and reproducibility checks instead — this skill explains that alternative).

neverlost-ai Bundle

Capture and completeness-check a supplied, bounded task for deferred execution while preserving scope, settled decisions, do-not-reopen constraints, inputs, blockers, and independent human-approval and capacity gates. Use when the user says to intake, queue, log, or save work for later without executing it. Do not use to perform the deferred task.

neverlost-ai Bundle

Apply the established Neverlost governed review process to documents, reports, PDFs, packets, presentations, manuscripts, context exports, templates, and workflow outputs. Use when reviewing, revising, cleaning, versioning, checkpointing, approving, formatting, visually QA-ing, or preparing a Neverlost artifact; when the user asks for a boundary review, evidence review, substantive review, review-only draft, production lock, acceptance test, operational approval, Governance Mode, or Execution Mode; or when a non-Neverlost artifact should be handled through the user's Neverlost review method.

nana7536 Skill

每天早上9点自动扫描吴双(Christina)飞书邮箱过去24小时收到的邮件，从中提取 KOL adcode 信息 （Post URL + Adcode + 产品），匹配到 Feishu Base 产品库，把完整记录连同 Month/Deliverables/ Posted?/Posting Date 一起写入 Content Log 子表单，并把格式化清单推送到吴双的飞书私信。也支持 吴双随时手动粘贴邮件原文/原始 IG/TikTok 链接触发同样的提取流程，当场在对话里回复。触发词： "adcode"、"今天的adcode"、"扫一下adcode邮件"、"帮我记录这些adcode"、"这是今天的红人链接"、 "extract adcode"、"处理一下这批adcode"，以及 cron job "Daily adcode email scan"。