713 个 · 按相关度排序
Add Agent Swarm (Teams) support to Telegram. Each subagent gets its own bot identity in the group. Requires Telegram channel to be set up first (use /add-telegram). Triggers on "agent swarm", "agent teams telegram", "telegram swarm", "bot pool".
中文简介正在准备中。当前显示原文说明。
Update documentation to match code changes. Triggered by: 'update docs', 'sync documentation', 'README outdated', 'docs更新', 'ドキュメント直して'. Delegates to subagent to protect context.
中文简介正在准备中。当前显示原文说明。
Multi-agent collaboration plugin that spawns N parallel subagents competing on the same task via git worktree isolation. Agents work independently, results are evaluated by metric or LLM judge, and the best branch is merged. Use when: user wants multiple approaches tried in parallel — code optimi...
中文简介正在准备中。当前显示原文说明。
Create custom AI subagents with proper plugin structure, persona generation, and companion routing skills.
中文简介正在准备中。当前显示原文说明。
Standard Operating Procedures and router for MCP task triage, subagent selection, and orchestration.
中文简介正在准备中。当前显示原文说明。
ADHD-friendly output shaping for subagent responses. Lead with the next action, number multi-step tasks, end with one concrete next step, suppress tangents, restate state across turns, give specific time estimates, make wins visible, matter-of-fact errors, cap visible lists to 5, no preamble or closers. Loaded by genin, jonin, anbu, tokubetsu-jonin, and chunin when shaping final responses.
中文简介正在准备中。当前显示原文说明。
Coordinate Codex work across the current task, internal subagents, user-owned tasks, worktrees, handoffs, formal review gates, or recurring automation. Use for delegation, parallel roles, another Codex task, cross-repo work, QA/review handoffs, scheduled follow-up, or task inspection and continuation. Chinese triggers include 子Agent/子进程/子对话, 另一个对话接手, 并行处理, 分给QA, 多仓库收口, 巡检, 持续推进, 定时自动继续, 创建/读取/转交任务.
中文简介正在准备中。当前显示原文说明。
Use when building, debugging, reviewing, migrating, or documenting Flue projects, including agents, hooks, tools, skills, subagents, sandboxes, routing, channels, persistence, schedules, observability, React clients, deployment targets, and CLI usage. Find and read version-matched Flue documentation before relying on API details.
中文简介正在准备中。当前显示原文说明。
Use for AI SDK for Python async-generator tools, streaming tool output, subagent tools, PartialToolCallResult events, and custom tool aggregation.
中文简介正在准备中。当前显示原文说明。
Procedimiento estructurado de 6 pasos para mejorar, renovar o reconstruir pipelines existentes, carpetas de proyectos individuales, estructuras de documentación o stacks de software. Direccionable como "pipeline optimizer" (para pipelines temáticos completos, p. ej. de software, investigación o desarrollo de videojuegos) o "project-folder optimizer" (para carpetas de proyectos individuales dentro de un pipeline, p. ej. una herramienta de software o proyecto de artículo académico). Se activa ante tareas como "mejorar pipeline X", "optimizar el stack", "reconstruir Y", "renovación", "refactorización de pipeline", "limpiar carpeta de proyecto", "mejorar estructura de carpetas", "unificar convenciones", "consolidación de documentación", "integrar en sistema existente" o cualquier intervención sustancial en estructuras establecidas. Proporciona análisis del estado actual, aclaración del propósito, boceto del estado ideal, plan de brechas, identificación empírica de puntos de dolor y reevaluación con subagentes limpios. Previene estándares paralelos, duplicación y rupturas en el pipeline.
中文简介正在准备中。当前显示原文说明。
Extrahiert aus einem Chatverlauf (aktuelle Session oder Transkript-Dateien) einen wiederverwendbaren Skill — oder verbessert einen sehr ähnlichen existierenden Skill, statt ein Duplikat zu erzeugen. Nutze diesen Skill bei „mach daraus einen Skill", „das sollten wir als Skill festhalten", „extrahiere Skills aus diesem/alten Chatverläufen", „diese Arbeitsweise wiederverwendbar machen", oder bei `/skill-extract`. Deckt auch Bulk-Läufe über viele alte Transkripte ab (mit Datenreduktion über Subagenten). Für wiederkehrende AUTOMATISIERUNGEN (Cron/Schedule/Loop) stattdessen den Schwester-Skill workflow-extract nutzen.
中文简介正在准备中。当前显示原文说明。
Structured 6-step procedure for improving, renovating, or rebuilding existing pipelines, individual project folders, documentation structures, or software stacks. Addressable as "pipeline optimizer" (for whole topic pipelines, e.g. a software, research, or game-dev pipeline) or "project-folder optimizer" (for individual project folders within a pipeline, e.g. a single software tool or paper project). Triggers on tasks like "improve pipeline X", "optimize the stack", "rebuild Y", "renovation", "pipeline refactoring", "clean up project folder", "improve folder structure", "unify conventions", "documentation consolidation", "integrate into existing system", or any substantial intervention in established structures. Delivers building-stock analysis, purpose clarification, ideal sketch, gap plan, empirical pain-point identification, and retests with fresh subagents. Prevents parallel standards, duplication, and pipeline breaks.
中文简介正在准备中。当前显示原文说明。
Audit BMAD source files for file-reference convention violations using parallel Haiku subagents. Use when users requests an "audit file references" for a skill, workflow or task.
中文简介正在准备中。当前显示原文说明。
Create custom AI subagents with proper plugin structure, persona generation, and companion routing skills.
中文简介正在准备中。当前显示原文说明。
Use when a user asks Claude Code to recommend, install, update, or inspect specialized coding subagents for Claude Code, Codex, OpenCode, or multiple targets in the current repository.
中文简介正在准备中。当前显示原文说明。
Use when a user asks Codex to recommend, install, update, or inspect specialized coding subagents for Claude Code, Codex, OpenCode, or multiple targets in the current repository.
中文简介正在准备中。当前显示原文说明。
Create a new Hindsight-powered subagent with long-term memory. Use when the user wants a specialized agent that learns and remembers across sessions.
中文简介正在准备中。当前显示原文说明。
Report AI token usage and estimated costs. Use when: owner asks about costs today/yesterday/this week, per session, or per model. Shows main session, cron jobs, and subagents. Answers: 'how much did today cost?', 'how much was this session?', 'what was last week's spend?'
中文简介正在准备中。当前显示原文说明。
This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent colors", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins.
中文简介正在准备中。当前显示原文说明。
Autonomously improve a real artifact (code, training recipe, agent harness, data pipeline, prompt) against an objective and an evaluator, using Hypothesis Tree Refinement (HTR) from the Arbor paper. Use this whenever someone wants to iteratively optimize something over many experiments without overfitting — e.g. "get my model's eval score up", "improve this agent/harness", "tune this pipeline", "beat the baseline on this benchmark", "run a search over approaches and keep the best", "do an MLE-bench / Kaggle-style optimization", or any long-horizon "make this artifact better and don't just memorize the dev set" task. Trigger it even when the user doesn't say "Arbor" or "hypothesis tree" but describes repeated experiment-and-evaluate loops, branching exploration of competing ideas, or worries about a dev/test gap. Runs Claude itself as the coordinator with subagent executors in isolated git worktrees; for the standalone `arbor` CLI tool see references/arbor-upstream.md.
中文简介正在准备中。当前显示原文说明。
TDD for process documentation - test with subagents before writing, iterate until bulletproof
中文简介正在准备中。当前显示原文说明。
Dispatch code-reviewer subagent to review implementation against plan or requirements before proceeding
中文简介正在准备中。当前显示原文说明。
Discover, validate, and invoke .agent.md custom agents. Use when finding agent files in the local Claude or VS Code Insiders directories, checking frontmatter, verifying disable-model-invocation, or determining agentName for runSubagent calls.
中文简介正在准备中。当前显示原文说明。
Use when facing 2+ independent tasks — "in parallel", "independent tasks", "multiple unrelated failures", "different test files or subsystems" — that can be worked on without shared state or sequential dependencies. Dispatch one agent per independent slice, each with only the context it needs, even when the user did not ask for subagents explicitly.
中文简介正在准备中。当前显示原文说明。