Context & AI engineering
25,745 rows, by stacks then stars
1,865 of these skill files have been read, by 1,328 distinct authors, and what they tell an agent counted. What Context & AI engineering authors agree on, and what they forbid
turntuptechnologies-ai/skills/skills/handoff Skill
1★ repoセッションの作業状態を引き継ぎ書として保存・復元するとき。「/clear する前に状態を保存」「引き継ぎ書を書いて」「今日はここまで」「さっきの続きから再開」「昨日の続き」等で使う。作業を途中で中断するとき・日をまたぐとき・コンテキストが逼迫してきた時・/clear・/compact の前にも自動で提案する。作業のゴール・進捗・重要ファイル・決定事項を `.claude/handoff/latest.md` に書き出し、新セッションで読み込んで作業を再開できるようにする。
turntuptechnologies-ai/skills/skills/library-eval Skill
1★ repo依存ライブラリやツールを選ぶとき。「どのライブラリを使うべき?」「○○ と △△ を比較して」「この依存を入れて大丈夫か調べて」等で使う。候補を保守状況・採用度・ライセンス・脆弱性・API 適合で比較し、推奨とリスクを出す。成熟した定番ライブラリを優先する方針に沿う。
turntuptechnologies-ai/skills/skills/run-agent-team Skill
1★ repoコード変更を伴う Issue(新機能・既存機能の変更・バグ修正)を Agent Teams で進めるとき。「チームで進めて」「Agent Teams でやって」等で使う。architect→reviewer→developer→tester→documenter を編成し、タスク依存で順に進め、解散前チェックまで回す。TeamCreate が使えない環境では確認の上 Subagents で代替できる。typo や 1 行の設定変更など軽微な変更はチーム不要。
gptme/gptme-cc-plugin/skills/gptme-run Skill
1★ repoDelegate a multi-step task to gptme, a terminal-based AI agent with its own tools (shell, files, browser). Use when the user invokes /gptme:run or asks to have gptme work on something autonomously using its own context and tools. Do NOT use for simple questions that Claude Code can answer directly.
broneq/bdk/skills/audit-prompt Skill
1★ repoAudit a system prompt or skill file against agent-architecture best practices. Returns scored findings; rewrite is opt-in.
broneq/bdk/skills/subagent-execute-plan Skill
1★ repoBDK — Broneq Dev Kit. Reusable Claude Code workflows: skills, agents, and hooks for TDD, planning, code review, and architecture documentation.
anshmajumdar121/skill-x/skills/super-skill Skill
1★ repoApply the combined rigor + delegation framework for complex tasks. Two operating modes: (1) **Rigor mode** — use the GPT-5.6 Sol 7-step loop, 10 principles, 9-stage architecture, task classification, risk register, validation layers, acceptance/rejection criteria, adversarial review, and pre-delivery checklist (default for any non-trivial task); (2) **Delegation mode** — for codebase-heavy or token-heavy work that can be split into independent parallel slices, use the Efficient Fable delegation pattern: cheap subagents do the heavy lifting, the orchestrator keeps judgment, architecture, synthesis, and final review; always apply the vetting protocol (treat subagent reports as leads, not facts). Trigger when the user says "super skill", "use the super framework", "delegate this efficiently", or any complex task where you want both auditable artifacts AND cost/speed efficiency. Do NOT trigger for trivial single-step questions, or for tasks that don't tolerate any subagent delegation (legal, medical, single-source-of-truth). This skill is **self-evolving**: it can add, update, or remove its own content when triggered by `evolve this skill`, when the agent notices stale/garbage content, or when a clearly better technique becomes available. Self-evolution is bounded by the constitution in `references/self-evolution.md` §1 (evidence-based, bounded scope, conservative, reversible, quality floor, no-garbage filter, privacy).
anshmajumdar121/skill-x/skills/think-like-gpt-5-6 Skill
1★ repoApply the GPT-5.6 Sol execution-intelligence framework to any complex task. Use when the user asks the agent to "think like GPT-5.6", "use the blueprint", "apply the execution framework", or wants auditable, requirement-driven, evidence-aware task execution. Also trigger on requests that benefit from a controlled 7-step loop (Observe → Interpret → Decide → Act → Verify → Repair → Record) with explicit acceptance criteria, validation gates, risk register, and a final pre-delivery checklist. Best fit for: multi-step coding, research with citations, artifact creation, operational actions, high-stakes guidance, and any task where the user wants inspectable reasoning rather than a fluent black-box answer. Do NOT trigger for trivial single-step requests where the loop overhead exceeds value, or when the user explicitly wants a fast informal answer.
bitcoin21ideas/skills/skills/plan-clash Skill
1★ repoHarden an existing implementation plan via a fixed N-round cross-model critique→fold loop. A codex (GPT) subprocess critiques the plan against the real code; a claude (opus) subprocess rules on each finding and folds accepted fixes into a copy. Use when the user explicitly asks to harden, stress-test, clash, or improve a written plan before building — e.g. "plan-clash this", "harden this plan". It improves a plan; it never certifies one — the human judges "good enough".
scoobydont-666/shared-claude-skills/skills/claude-to-cursor Skill
1★ repoConvert Claude Code skills to Cursor-compatible rules (.mdc/.md files) for sharing or use in Cursor workspaces. Does NOT modify local Claude Code setup — outputs go to a separate repo. Trigger on: "convert to cursor", "cursor rules", "port to cursor", ".mdc", "cursor format", or any request to export skills for Cursor users.
scoobydont-666/shared-claude-skills/skills/config-auditor Skill
1★ repoCross-reference all CLAUDE.md files, memory files, skill descriptions, and project registries for contradictions, stale data, and inconsistencies. Trigger on "check for contradictions", "audit configs", "config consistency", "are my configs consistent", or any request to verify configuration coherence across Project Swarm.
scoobydont-666/shared-claude-skills/skills/hydra-swarm Skill
1★ repoDistributed Claude Code coordination v2 — autonomous multi-instance orchestration via NFS + git. Agent registry with heartbeat, priority work queue with capability matching, event bus for cross-agent context, auto git sync, GPU slot management, session lifecycle protocol. Replaces OpenClaw for self-hosted agent coordination.
scoobydont-666/shared-claude-skills/skills/token-miser Skill
1★ repoMinimize Claude token and cost waste across ALL surfaces — Claude Code subagents, API calls in code, and artifacts. Routes subagent model selection via the Agent tool's model parameter. Recommends /model switches when the current model is overkill or underpowered. Applies to any code importing anthropic or calling /v1/messages. Trigger on: "API call", "artifact", "pipeline", "multi-agent", "LangGraph", "which model", "save tokens", "reduce cost", "claude API", "token usage", "context window", any code importing anthropic or calling /v1/messages, subagent launches, or any discussion about optimizing Claude usage or costs.
RealDougEubanks/ClaudeMarketplace/skills/agent-based-development/skills/agent-based-development Skill
1★ repoFull async multi-agent development workflow: Planning → Design → Dev → Security/Tech Review loop with file-based handoffs and release-branch Git model.
darbin/claudecraft/plugins/rem-meta/skills/rem-feedback Skill
1★ repoCapture typed feedback about skill mistakes so skills can learn and improve over time. Writes structured entries to `~/.claude/skill-feedback.jsonl` tagged with one of six mistake types — misrouted (wrong skill fired / should have fired), wrong-output (false claim, bad advice), missed (skill didn't flag an issue it should have), over-flagged (skill flagged a non-issue), context-ignored (skill ignored CLAUDE.md / memory / project conventions), stale (advice references deleted files / old APIs / outdated facts). Links each feedback entry back to the originating invocation in `~/.claude/skill-usage.jsonl` by session_id + timestamp. Auto-promotes to `/rem-skill analyze [name]` when 3+ same-type flags accumulate for one skill. Distinct from rem-learn (which captures project-level lessons, not skill-system lessons) and rem-handoff (session state, not feedback). Use for "flag a skill mistake", "this skill got it wrong", "the skill should have X", "rem-copy misrouted", "rem-review missed the bug", "skill feedback", "rate the skill", "this wasn't helpful", after any skill-assisted session where output disappointed.
darbin/claudecraft/plugins/rem-meta/skills/rem-handoff Skill
1★ repoSession continuity and handoff. Saves current session state for future resumption, or resumes from a previous handoff. Use for handoff, session handoff, context transfer, resume, continue, pick up where I left off, save progress, session state, or context preservation.
darbin/claudecraft/plugins/rem-meta/skills/rem-learn Skill
1★ repoExtract permanent lessons from the current session so they're never repeated. Mines git diff / reflog / conversation for error→fix chains, silent failures (code ran but produced wrong data), near-misses (almost deployed X), wasted effort (spent 30m on wrong approach), positive patterns worth replicating, and architecture insights uncovered during debugging. Writes grep-able entries to learnings.md / auto-memory / CLAUDE.md with code, exact error messages, dates, keywords — NOT "be more careful with types". Escalates 3+ similar entries to project CLAUDE.md conventions, universal lessons to global `~/.claude/CLAUDE.md`. Prunes stale entries, keeps MEMORY.md under 200 lines. Distinct from rem-handoff (which saves session-resumption state, not lessons). Use for "learn from this", "remember this", "update learnings", "never do this again", "capture lessons", "save what we learned", or after any debugging / error session.
darbin/claudecraft/plugins/rem-meta/skills/rem-loop Skill
1★ repoAutonomous agentic loop patterns for multi-iteration work — PR Loop (branch → implement → CI → auto-fix → repeat with SHARED_TASK_NOTES.md context bridge) and De-Sloppify (two-pass - implementation agent then dedicated cleanup agent). Use when a task requires many iterations against CI feedback, when you want a post-implementation cleanup pass without negative constraints, or when you need to bridge context between loop iterations. Distinct from the built-in /loop (which schedules recurring prompts on an interval) — rem-loop provides the PATTERNS for building autonomous development loops. Use for "autonomous loop", "keep running until CI passes", "pr loop", "loop until green", "agentic loop", "cleanup pass", "de-sloppify", "two-agent pattern", "context bridge", "SHARED_TASK_NOTES".
darbin/claudecraft/plugins/rem-meta/skills/rem-skill Skill
1★ repoBuild, evaluate, improve, and analyze Claude Code skills. Encodes skill-design expertise — description optimization, progressive disclosure via `_references/`, knowledge-delta scoring, archetype selection, anti-pattern detection. ANALYZE mode reads `~/.claude/skill-feedback.jsonl` (populated by `/rem-feedback`), aggregates by skill + mistake_type, and proposes targeted edits — description triggers for misrouted flags, new anti-patterns for missed, severity calibration for over-flagged, Step-0-context for context-ignored, freshness stamps for stale. IMPROVE mode consumes that analysis and applies edits, marking entries `addressed`. Use when the user says "create skill", "new skill", "evaluate skill", "audit skill", "improve skill", "refactor skill", "skill builder", "meta-skill", "analyze skill", "aggregate feedback", "why is rem-X misfiring", "skill report card", "skill health", "skill won't load", "skill not triggering", "skill not firing", "diagnose skill", "skill doctor", "why isn't my skill loading".
ZLHad/agent-skills/claude-code-skills/knowledge-base-manager Skill
1★ repo智能管理和整理本地知识库文档。扫描、分析、整合由conversation-knowledge-extractor生成的知识库文件,生成结构化索引、专题整理报告、进步追踪分析。支持增量/全量整理,按日期范围筛选。重点处理学术写作和学术探讨内容(生成错误手册、模式库、思维框架库),同时支持其他类型文档的基础整理。触发词:"整理知识库"、"管理本地知识"、"生成学习报告"、"更新知识索引"等。
ZLHad/agent-skills/openclaw-skills/claude-code-bridge Skill
1★ repoBridges OpenClaw (QQ, Telegram, WeChat, and other messaging channels) to a persistent Claude Code CLI session running in a background tmux process. Enables starting, stopping, restarting, and monitoring Claude Code sessions directly from any chat interface. Supports specifying a working directory or launching in sandbox mode (temp directory, auto-cleanup on stop). Automatically detects session state on every message, routes user input to the active Claude Code session, and handles tool-approval prompts so the user can approve or deny Claude Code actions without leaving their chat app. Trigger phrases: "start claude code", "open claude code", "cc status", "stop claude code", "restart cc", "启动claude code", "开启claude code", "启动cc", "开启cc", "连接cc", "cc状态", "关闭cc", "退出cc", "重启cc", "在...打开cc", "沙盒打开cc", "沙盒模式启动cc", "/cc start", "/cc stop", "/cc restart", "/cc status".
goldenprofile/llm-skills/agent-audit Skill
1★ repoСамоаудит AI-агента (не аудит кода): проверка актуальности проектной документации, поиск повторяющихся ошибок агента, генерация guardrails и конкретных правок в файлы проекта. Используй когда пользователь говорит «проверь себя», «самоаудит», «self-check», «мета-ревью», «что устарело в CLAUDE.md», «обнови контекст», «что ты упускаешь», «какие ошибки повторяешь», «насколько хорошо ты понимаешь проект», или после крупной работы («мы закончили фичу», «спринт завершён», «что дальше»). НЕ триггерится на мелких фиксах; для аудита кода или проекта — см. django-audit / python-project-audit.
goldenprofile/llm-skills/clarify-prompt Skill
1★ repoПревращает нечёткие, разговорные описания задач в структурированные, однозначные промты для AI-агента (Claude Code, Claude Chat и др.). Используй когда пользователь просит «улучши промт», «перепиши задачу для ИИ», «структурируй задачу», «clarify prompt», «напиши ТЗ для агента», «сформулируй задачу», упоминает prompt engineering, или даёт сырой текст задачи и хочет получить чистую инструкцию. Работает с любым языком ввода.
goldenprofile/llm-skills/claude-code-auditor Skill
1★ repoАудит кода, написанного Claude Code, через «второе мнение» другой LLM (DeepSeek/GLM/Kimi): вызывает CLI Hermes (`hermes -z`) — свежий взгляд не-Claude модели: галлюцинации API, логические ошибки, edge cases, over-engineering. Пять линз: Скептик, Безопасность, Надёжность, Производительность, Поддерживаемость; режимы — коммит(ы), файлы, архитектура. Используй когда пользователь просит «второе мнение», «second opinion», «ревью от другой модели», «проверь что написал Claude», «линза скептика». Требует CLI `hermes`. Для ревью силами самого Claude — techlead-ai.
goldenprofile/llm-skills/context-hygiene Skill
1★ repoГигиена контекста Claude Code: диагностика деградации длинной сессии («агент забыл», повторные вопросы, правки не туда), решение «продолжать или новая сессия», разгрузка контекста (разведка сабагентами, итоги в файлы, /compact в осмысленных точках) и раскладка постоянного знания по слоям — CLAUDE.md (правила), docs/ (устройство), навыки (процедуры), память агента (факты). Включает чек-лист качества CLAUDE.md. Используй когда пользователь спрашивает «что положить в CLAUDE.md», «новая сессия или продолжаем», «почему ты потерял контекст», «оптимизируй контекст», жалуется что агент забывает договорённости, или сессия распухла и агент деградирует. Самоаудит повторяющихся ошибок агента и актуальности документации — agent-audit; восстановление контекста после перерыва — session-catchup.
goldenprofile/llm-skills/docs-generator Skill
1★ repoГенерация и аудит документации для соло-разработчика: README, ADR, Python-docstrings (Google-стиль), синхронизация CLAUDE.md/AGENTS.md. Документирует «почему», а не пересказывает код; находит отсутствующую и устаревшую документацию. Используй когда пользователь просит написать или обновить README, оформить ADR, добавить или проверить docstrings, синхронизировать CLAUDE.md и AGENTS.md, или говорит «задокументируй проект», «doc audit». Для спек/планов/брифов — см. spec-writer; DoD/tooling-обвязку агентских правил настраивает harness-engineering.
goldenprofile/llm-skills/harness-engineering Skill
1★ repoВнедрение harness engineering (обвязки для AI-агентов) в Python-проект соло-разработчика: Django, FastAPI, aiogram. Создаёт Makefile, CI (GitHub Actions), ARCHITECTURE.md, обновляет CLAUDE.md/AGENTS.md (DoD, tooling, canonical docs) и вшивает в Definition of Done вызовы навыков библиотеки (migration-safety-auditor, python-project-audit, test-coverage-auditor) и официальных гейтов (/code-review, /security-review, pyright-lsp). Деплой-ориентир — systemd/nginx/redis/postgres, не Docker. Используй когда пользователь просит настроить harness, подготовить проект для агентов, внедрить DoD/tooling-обвязку, или говорит «harness», «symphony», «оркестрация агентов».
goldenprofile/llm-skills/llm-delegation Skill
1★ repoРешение «делать самому или поручить агенту» и как именно: карта сильных и слабых сторон LLM в разработке, матрица делегируемости (проверяемость результата × формализуемость контекста), выбор режима Claude Code под тип задачи (прямой заход, plan mode, сабагенты, /goal, петли) и уровня модели/усилия. Плюс разбор пост-фактум: почему агент справился плохо и каким навыком это чинить. Используй когда пользователь спрашивает «стоит ли это делегировать ИИ», «сам или агент», «какую модель/режим выбрать», «почему агент плохо справился», планирует распределение работы между собой и агентом, или систематически получает плохие результаты на определённом классе задач. Декомпозиция уже принятой в работу задачи — task-decomposer; формулировка промта — clarify-prompt.
goldenprofile/llm-skills/llm-feature-architect Skill
1★ repoПроектирование и ревью LLM-фич в собственном продукте (Python: Django/FastAPI/aiogram): нужен ли вообще LLM, форма задачи (классификация/извлечение/генерация/диалог/агент) и выбор модели, каркас интеграции (сервис-обёртка, таймауты, ретраи с backoff, очередь для долгих вызовов), structured output через tool use + pydantic, контроль стоимости (prompt caching, max_tokens, метрики на пользователя), evals на золотом наборе, безопасность (prompt injection, PII). Используй когда пользователь добавляет ИИ-фичу в приложение, вызывает LLM API из кода, спрашивает «как прикрутить Claude/GPT к продукту», проектирует суммаризацию, классификацию, извлечение данных или чат-бота, жалуется на стоимость вызовов или нестабильный JSON от модели. Актуальные ID моделей и цены Anthropic — навык claude-api; выбор «делать самому или агентом» в разработке — llm-delegation.
Asm3r96/wave-driven-dev/skills/wave-planner-claude Skill
1★ repoPlan and execute large multi-step, multi-stack changes with contract-first waves, parallel worker ownership, user approval gates, ordered integration, and final verification/docs updates using Claude Code Agent subagents. Use only when the task spans multiple areas (for example frontend, backend, tests, errors, docs) or has dependency-heavy sequencing.
donatassimkus/claude-ai-skills/skills/claude-code-audit Skill
1★ repoAUDIT: is the Claude Code rig itself healthy? Audit and self-maintain the Claude Code setup. Auto-updates the capabilities reference, diagnoses what's working across memory/skills/hooks/rules/security, auto-executes safe cleanup and batches the rest for approval. Use when asked to audit the setup, optimize workflows, review Claude Code health, find automation opportunities, or clean up the system.
donatassimkus/claude-ai-skills/skills/handoff Skill
1★ repoHand off the current conversation to a fresh chat with clean tokens. Synthesizes the session into a compact brief file and a paste-ready prompt so work continues without losing context. Use when the chat is token-heavy (judged as a fraction of this platform's context window, not an absolute count), when response quality is degrading from bloat, or when the user wants to switch to a fresh session without losing the thread.
donatassimkus/claude-ai-skills/skills/portable-prompt Skill
1★ repoGrowth marketing skills for Claude Code, Cursor, and any AI agent: SEO, CRM, marketing automation, inbox triage, and pre-launch QA. Paste one prompt and your agent installs the method, then applies it to your own accounts.
lennney/gate-all-skills/skills/design/product-analysis Skill
no license1★ repoMulti-path parallel product analysis with cross-model test-time compute scaling. Spawns parallel agents (Claude Code agent teams + Codex CLI) to explore product from multiple perspectives, then synthesizes findings into actionable optimization plans. Can invoke competitors-analysis for competitive benchmarking. Use when "product audit", "self-review", "发布前审查", "产品分析", "analyze our product", "UX audit", or "信息架构审计".
lennney/gate-all-skills/skills/discussion/handoff Skill
no license1★ repoCompact the current conversation into a handoff document for another agent to pick up.
YuAICode/ai-skills/skills/claude-code-zh Skill
1★ repo把 Claude Code 汉化 —— 让 Claude 默认用中文回复,并给每个工具调用加中文 tooltip 提示。当用户想"汉化 Claude Code""让 Claude 说中文""装中文命令提示""localize Claude Code to Chinese"时使用。
wasulajr/headsup/skills/nil Skill
1★ reponow is later: the companion to /sfl. Reopen the windows you saved for later: list the saved checkpoints in ~/.claude/sfl/ numbered, ask which to launch (all or specific numbers), then open one tab per selected entry in the current terminal provider, or AI Power Term when selected or when run from AI Power Term. Each tab cds to its dir, restores its label, and launches the current/forced LLM CLI (Claude or Codex) pointed at the archived checkpoint. Only launched entries are archived; un-selected ones stay live for a later /nil. Use when the user types /nil, "now is later", "reopen my windows", "restore my tabs", or "bring my saved windows back".
KirKruglov/OFFICINA/skills/merging-branches Skill
1★ repoUse BEFORE any git merge, rebase, or branch integration. Reads the project conventions file (CLAUDE.md or AGENTS.md) to determine merge strategy, runs pre-merge safety checks (diff name-status both sides, clean working tree, up-to-date with remote, conflict scan), requires explicit user confirmation for merge to main/master/release branches. Trigger phrases — "merge X into Y", "rebase onto", or any request implying branch integration.
im-shashanks/claude-plugins/shaktra/skills/shaktra-memory-stats Skill
no license1★ repoMemory system inspector. View learned principles, anti-patterns, procedures, per-story briefings, and seed new entries. Complements status-dash with detailed memory audit capabilities.
im-shashanks/claude-plugins/shaktra/skills/shaktra-memory Skill
no license1★ repoInternal skill for memory consolidation and retrieval — observation classification, confidence math, synthesis algorithms, and tiered briefing generation. Loaded by the memory-curator and memory-retriever agents.
Chrike/coding-agent-skills/skills/markdown-memory Skill
no license1★ repoUse when the user clearly asks to record, update, prune, or consult project-governed Markdown reference lessons about repeated mistakes, corrections, or confirmed approaches that must be versioned, shared, reviewable, or otherwise tied to repository history. Do not use for automatically loaded CLAUDE.md or .claude/rules instructions, host auto memory, handoff state, or decision-frontier planning.
Chrike/coding-agent-skills/skills/memory-handoff Skill
no license1★ repoCreates or updates a compact handoff or checkpoint for a later session, and resumes work from a user-named or repository-standard handoff artifact. Use when the user explicitly asks to create or update a handoff or checkpoint, prepare one before context compaction, or resume from one. Do not use for ordinary progress summaries, long contexts without explicit handoff intent, or a direct request to run /compact without preparing a handoff.
scagogogo/versions-skills/skills/mcp-operations Skill
1★ repoInvoke version operations via the versions MCP server — tool naming, parameter conventions, server configuration, and batch patterns.
Yesterday-AI/ytstack/skills/handoff-session Skill
1★ repoUser-triggered session handoff. Writes a detailed HANDOFF.md with current state, open items, pending decisions, and recommended resume prompt. Use before switching machines, before long breaks, or when you need to clear context intentionally. Parallel to the automatic pre-compact hook.
Yesterday-AI/ytstack/skills/spawn-milestone-team Skill
1★ repoDispatch the current milestone's slices to a Claude Code Agent Team. Each teammate works on a slice in parallel with its own 200k context (replaces GSD's fresh-subprocess-per-task pattern). Requires CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1 and v2.1.32+. Run after slice-milestone + plan-eng-review.
Yesterday-AI/ytstack/vendor/gstack/benchmark-models Skill
1★ repoCross-model benchmark for gstack skills. Runs the same prompt through Claude, GPT (via Codex CLI), and Gemini side-by-side — compares latency, tokens, cost, and optionally quality via LLM judge. Answers "which model is actually best for this skill?" with data instead of vibes. Separate from /benchmark, which measures web page performance. Use when: "benchmark models", "compare models", "which model is best for X", "cross-model comparison", "model shootout". (gstack) Voice triggers (speech-to-text aliases): "compare models", "model shootout", "which model is best".
Yesterday-AI/ytstack/vendor/gstack/benchmark Skill
1★ repoPerformance regression detection using the browse daemon. Establishes baselines for page load times, Core Web Vitals, and resource sizes. Compares before/after on every PR. Tracks performance trends over time. Use when: "performance", "benchmark", "page speed", "lighthouse", "web vitals", "bundle size", "load time". (gstack) Voice triggers (speech-to-text aliases): "speed test", "check performance".
Yesterday-AI/ytstack/vendor/gstack/pair-agent Skill
1★ repoPair a remote AI agent with your browser. One command generates a setup key and prints instructions the other agent can follow to connect. Works with OpenClaw, Hermes, Codex, Cursor, or any agent that can make HTTP requests. The remote agent gets its own tab with scoped access (read+write by default, admin on request). Use when asked to "pair agent", "connect agent", "share browser", "remote browser", "let another agent use my browser", or "give browser access". (gstack) Voice triggers (speech-to-text aliases): "pair agent", "connect agent", "share my browser", "remote browser access".
Yesterday-AI/ytstack/vendor/superpowers/skills/dispatching-parallel-agents Skill
1★ repoUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
The same file is in 13 other repositories
Yesterday-AI/ytstack/vendor/superpowers/skills/subagent-driven-development Skill
1★ repoUse when executing implementation plans with independent tasks in the current session
The same file is in 6 other repositories
yokeloop/yoke/skills/handoff Skill
1★ repoSaves the live state of the current conversation to `.yoke/handoff/` so a fresh session resumes where this one stopped, referencing existing artifacts instead of duplicating them. Activates when the user writes "handoff", "hand off", "save the context", "dump the conversation", "compact the conversation", "summarize for the next session", "prepare a handoff doc", "pass this to another agent".
claude-hangar/claude-hangar/core/skills/handoff Skill
1★ repoStructured session handoff — preserves context for seamless continuation across sessions.
claude-hangar/claude-hangar/core/skills/hook-gen Skill
1★ repoGenerate Claude Code hook configurations from natural language descriptions. Use when: "create hook", "add hook", "hook for", "when X happens do Y", "hook-gen".
claude-hangar/claude-hangar/core/skills/prompt-optimizer Skill
1★ repoAnalyzes draft prompts, identifies gaps, matches available skills/agents, and recommends optimal workflows. Advisory only — never executes, only optimizes. Use when unsure which skills or approach to use for a task.
limerickgds/dashi-skills/skills/cybernetics-agent-seminar Skill
1★ repoA cybernetics × AI agent seminar skill. Grounded in *Cybernetics and Scientific Methodology* (Jin Guantao) and Wienerian cybernetics, it analyzes how cybernetic principles inform modern AI agent R&D, architecture review, and design guidance. Trigger when users mention cybernetics, feedback control, agent architecture, systems theory, black-box methods, negative feedback, positive feedback, agent design principles, steady-state systems, information entropy, purposive behavior, functional simulation, or Ashby's law of requisite variety. Also use when users want to review agent architectures from a cybernetic perspective, discuss feedback mechanism design, analyze agent stability and self-evolution, or organize cybernetics × AI agent seminars.
Zettelkasten knowledge gardener
limerickgds/dashi-skills/skills/zettelkasten-knowledge-gardener Skill
1★ repoA Zettelkasten-based knowledge gardening skill. Use this whenever users want to organize a knowledge base, 整理知识库, 整理读书笔记, build a second brain, update a personal wiki, turn conversations, reading notes, or project learnings into evergreen notes, MOCs, PARA folders, reusable context packs, or long-term memory assets. Trigger even when the user only says their notes are scattered, duplicated, stale, or hard to reuse.
odere-pro/claude-wiki-pages-plugin/skills/engine-api Skill
1★ repoThe LLM-facing contract for the deterministic claude-wiki-pages Bun engine — how ANY agent should call it. Documents each subcommand (verify, fix, heal, doctor, config, migrate, search, snapshot; plus planned index/link-suggest), its `--json` output shape, exit codes, and when to call it on a write-path. Trigger when an agent or user asks "how do I call the engine", "what does the engine return", "how do I verify/heal programmatically", or invokes /claude-wiki-pages:engine-api. This is reference material, not an action — it teaches the tool surface.
odere-pro/claude-wiki-pages-plugin/skills/ingest-pipeline Skill
1★ repoThe ingest pipeline's topic-tree plan format and confirmation gate, the opt-in destructive optimize (restructure) procedure, and the final-report template for claude-wiki-pages. Documents the plan/approval and restructure steps so the ingest agent can drive them without inlining the detail. Trigger when an agent or user asks "how does ingest plan the topic tree", "how does the optimize restructure work", or invokes /claude-wiki-pages:ingest-pipeline. Reference, not action.
odere-pro/claude-wiki-pages-plugin/skills/ingest Skill
1★ repoIngest one or more sources from vault/raw/ into typed wiki pages under vault/wiki/. Trigger when the user says "ingest this source", "process the file I just dropped in raw/", "add this to the wiki", or invokes /claude-wiki-pages:ingest directly. Prefer the pipeline (/claude-wiki-pages:claude-wiki-pages-ingest-agent) unless the user has asked to skip lint-fix and synthesis.
odere-pro/claude-wiki-pages-plugin/skills/init Skill
1★ repoOnboarding entry point. Scaffold a new LLM Wiki vault in the user's project by copying from the skill's own template/, stamp the schema version, and orient the user. Trigger when the user says "set up a wiki", "initialize the vault", "start a new LLM Wiki", "bootstrap the vault", or invokes /claude-wiki-pages:init directly.