Communitygithub.com

智能体技能自动发现与匹配

自动发现和匹配智能体技能,实现动态任务路由和能力评估。

智能体技能自动发现与匹配 是什么?

该技能专注于解决智能体在学习过程中的“鸡-蛋”问题:即如何在不知道成功标准的情况下学习成功,并同时定义成功标准。它通过架构分离“提议”和“验证”两个过程,避免循环依赖,从而构建一个既能自我教学又能自我评估的自主学习系统。 技能详细分析了不同场景下的架构选型,例如当任务有明确成功标准时使用单模型架构进行快速评估,而对于开放式发现任务则采用双过程架构(快速LLM评估结合独立VLM验证),或通过自上而下的任务分解来解决复杂问题。 技能还深入剖析了五种关键失败模式(如橡皮图章评估、基元爆炸等),并为每种模式提供了症状、检测方法和修复策略。通过丰富的案例(如机器人开门、煮咖啡、RAG环境知识蒸馏)展示了双过程验证、任务分解和知识蒸馏的实际应用效果。 该技能适用于Claude Code、Cursor、Codex等主流AI编程平台,旨在帮助开发者设计更鲁棒的自主学习和动态任务路由系统。

兼容平台~Claude Code~Codex CLI~Cursor✓Antigravity
npx skills add https://github.com/curiositech/windags-skills/tree/HEAD/skills/agentic-skill-discovery

在你喜欢的 AI 中提问

打开一个已预加载此 Agent Skill 的新对话。

文档

智能体技能自动发现

版本: 1.0 领域: 自主学习系统、AI架构、强化学习

核心概念

自主技能发现解决了鸡-蛋问题:在学习如何成功的同时学习什么构成成功。系统必须既是学生又是老师,这造成了需要架构上分离提议和验证的循环依赖。

决策点

架构选择决策树

如果任务具有真实成功标准(例如,具有明确目标的导航、操作任务) → 使用单模型架构进行快速LLM评估 → 成本:低 | 精度:~75% | 用例:预定义任务

如果任务需要开放式技能发现(例如,创意任务、探索) → 使用双过程架构(系统1/系统2) → 系统1:用于训练循环的快速LLM评估 → 系统2:用于技能库准入的独立VLM验证 → 成本:高 | 精度:~76% | 用例:自主学习

如果需要复杂的多步骤行为 → 自上而下的任务分解 → 从失败复杂任务开始 → 分解为子任务 → 学习缺失的前提条件 → 成功率:43.75% vs 12.50%(自下而上链接)

如果从原语构建技能库 → 仅当原语与最终目标匹配良好时,才使用自下而上的技能链接 → 风险:组合爆炸和缺失“中间技能” → 备用方案:当组合失败时切换为自上而下

如果现有方法性能停滞 → 检查循环依赖(同一模型提议和评估) → 如果检测到:实现架构分离 → 如果未检测到:添加包含成功技能模式的RAG,用于环境知识蒸馏

验证策略决策表

场景快速评估(系统1)慢速评估(系统2)技能库准入规则
训练循环由LLM生成的成功函数无不适用
技能验证LLM评估VLM验证要求两者均通过
技能库污染风险高容忍度零容忍度需要系统2覆盖
成本约束优化速度优化精度不对称成本接受

失败模式

1. 橡皮图章评估

症状: 报告成功率很高但任务完成度差,技能库快速增长 检测规则: 如果同一模型既生成奖励又评估成功,则存在循环依赖 根本原因: LLM同时充当球员和裁判会产生评估偏差 修复: 实现架构分离,使用独立验证器(例如,视觉任务使用VLM)

2. 基元爆炸瘫痪

症状: 技能库庞大但无法解决复杂任务,技能无法有效组合 检测规则: 如果技能数量增长但复杂任务成功率停滞,则存在“中间技能”问题 根本原因: 自下而上的链接无法发现介于基元和复杂目标之间的技能 修复: 从期望的复杂行为切换为自上而下的任务分解

3. 无上下文的RAG浪费

症状: RAG检索发生但性能没有提升,将检索视为通用示例 检测规则: 如果RAG没有改进约束学习(例如,物理理解),则只是昂贵的提示工程 根本原因: 缺少环境知识蒸馏机制 修复: 将技能库结构化为环境可供性的渐进模型,而不仅仅是例子的集合

4. 技能库污染级联

症状: 随着技能库增长,性能逐渐下降,假阳性不断累积 检测规则: 如果技能库准入使用与训练循环相同的评估,污染将会级联 根本原因: 糟糕的技能使得糟糕的组合成为可能;技能库错误会累积,不同于训练噪声 修复: 技能库准入使用昂贵的独立验证,训练循环仅使用廉价评估

5. 手动粒度覆盖

症状: 技能感觉与自然语义边界脱节,抽象级别强制指定 检测规则: 如果手动指定技能粒度与LLM提议矛盾,检查领域对齐 根本原因: LLM从训练数据中编码了语义任务边界 修复: 让系统自由提议技能,观察发现的粒度,仅针对特定领域需求进行调整

工作实例

实例1:使用双过程验证的机器人抽屉打开

场景: 智能体必须学习打开各种抽屉类型,且没有预定义的成功标准。

系统1(快速)过程:

  1. LLM生成技能提议:"抓取抽屉把手"
  2. LLM创建成功函数:drawer_handle_grasped = gripper_distance < 0.02
  3. RL训练使用该函数进行100个回合
  4. LLM评估:"成功!夹爪始终能到达把手"
  5. 决策点: 是否加入技能库?否 - 需要系统2

系统2(慢速)过程:

  1. VLM分析来自10个最佳回合的最终视频帧
  2. VLM判断:"夹爪接触把手但未保持抓握 - 手指未闭合"
  3. 检测: 捕获系统1假阳性
  4. 动作: 拒绝技能库准入,完善成功函数
  5. 迭代: 在成功函数中添加抓握压力要求

专家与新手对比:

  • 新手遗漏: 信任LLM自我评估,收录学习不佳的技能
  • 专家捕获: 同一模型无法评估自己提议的内容;需要独立验证

结果: 技能库保持76%的精度,而未经系统2验证时仅为46%

实例2:多步骤任务的自上而下任务分解

场景: 使用自下而上的技能链接执行"准备咖啡"失败,尽管拥有诸如"抓取杯子"、"倒液体"等原语

自下而上失败分析:

  • 可用技能:抓取杯子、倒液体、移动手臂、检测物体
  • 组合尝试:尝试了所有16种可能的2技能链
  • 缺失的中间技能: "将对准杯子与壶嘴"、"检测液体液位"、"协调双臂倒液"
  • 决策点: 自下而上无法发现这些技能 - 它们不是原语组合

自上而下分解:

  1. 失败任务:"准备咖啡"
  2. 分解为:[加热水] → [研磨咖啡豆] → [对准杯子] → [倒水] → [加入咖啡豆]
  3. 差距分析: 技能库中缺失"对准杯子"
  4. 按需学习: 学习"对准杯子",初始状态 = "研磨咖啡豆"的结束状态
  5. 链接: 每个子任务的成功状态成为下一个任务的初始条件

权衡分析:

  • 自下而上:12.50%的成功率,探索不必要的技能组合
  • 自上而下:43.75%的成功率,仅学习必需的前提条件
  • 成本: 自上而下需要任务分解能力,自下而上需要穷举探索

自上而下何时会适得其反:

  • 如果任务分解错误,则为错误的子任务学习技能
  • 如果初始状态无法可靠设置,链接会中断
  • 缓解措施: 在技能学习前验证分解,实现状态设置验证

实例3:RAG环境知识蒸馏

场景: 智能体必须在新环境中学习操作技能,没有明确的物理约束

无RAG(25%成功率):

  1. LLM生成奖励:reward = distance_to_target
  2. 智能体学会接近但未考虑碰撞边界
  3. 失败: 不知道抓取前夹爪必须距离表面5厘米

有RAG(46%成功率):

  1. 检索之前的成功:"reach_cube_A"奖励函数
  2. 检测到的模式: 所有成功的抓取都包含gripper_height > 0.05约束
  3. LLM整合:reward = distance_to_target if gripper_height > 0.05 else -10
  4. 知识蒸馏: 隐式学习了环境可供性

专家与新手对比:

  • 新手遗漏: 将RAG视为示例提供,未认识到约束学习
  • 专家捕获: RAG通过模式识别实现环境知识蒸馏

RAG何时会适得其反:

  • 如果检索的模式来自不同环境,则学习到错误的约束
  • 如果成功示例过于稀疏,则模式不具统计显著性
  • 缓解措施: 按环境相似性过滤检索,要求最小样本量进行模式提取

质量门

  • 提议和验证使用不同的模型能力(例如,LLM提议,VLM验证)
  • 在留出测试集上测量,快速评估精度>45%,慢速评估精度>75%
  • 技能库准入需要独立验证,而不仅仅是训练循环的成功
  • 复杂任务分解生成3-7个有意义的子任务,并具有清晰的依赖顺序
  • RAG检索将约束学习提升超过无RAG基线的15%
  • 技能粒度分析显示语义聚类,而非参数枚举
  • 系统能够检测并从循环依赖失败模式中恢复
  • 通过独立评估验证,技能库污染率<25%
  • 在复杂任务上,自上而下分解的性能优于自下而上链接超过2倍
  • 环境知识蒸馏展示了从成功模式中进行约束学习的能力

非适用边界

请勿将本技能用于:

  • 具有已知成功标准的预定义任务 → 改用带有手工设计奖励的标准RL
  • 单步骤操作任务 → 改用带有演示数据的监督学习
  • 需要实时性能的任务 → 双过程验证太慢;改用缓存的技能库
  • 没有视觉反馈的领域 → VLM验证不可用;设计特定领域的独立验证器
  • 资源受限环境 → 系统2验证成本高;接受较低精度或使用分层验证

改用以下技能:

  • 标准奖励工程 → 适用于具有明确成功指标的任务
  • 模仿学习 → 适用于有大量演示数据的任务
  • 分层强化学习 → 适用于具有已知技能分解的任务
  • 元学习 → 适用于快速适应新的任务分布
  • 宪法式AI → 适用于需要价值对齐和安全约束的任务

Individual skills in this repo

This repo contains 20 individual skills — each has its own dedicated page.

curiositech/windags-skills

Expert in 2000s-era music visualization (Milkdrop, AVS, Geiss) and modern WebGL implementations. Specializes in Butterchurn integration, Web Audio API AnalyserNode FFT data, GLSL shaders for audio-reactive visuals, and psychedelic generative art. Activate on "Milkdrop", "music visualization", "WebGL visualizer", "Butterchurn", "audio reactive", "FFT visualization", "spectrum analyzer". NOT for simple bar charts/waveforms (use basic canvas), video editing, or non-audio visuals.

curiositech/windags-skills

Expert legal research agent for finding and scraping expungement data state by state. Knows authoritative sources, URL patterns, Firecrawl configuration, and 2026 legal landscape.

curiositech/windags-skills

Expert in 3D computer vision labeling tools, workflows, and AI-assisted annotation for LiDAR, point clouds, and sensor fusion. Covers SAM4D/Point-SAM, human-in-the-loop architectures, and vertical-specific training strategies. Activate on '3D labeling', 'point cloud annotation', 'LiDAR labeling', 'SAM 3D', 'SAM4D', 'sensor fusion annotation', '3D bounding box', 'semantic segmentation point cloud'. NOT for 2D image labeling (use clip-aware-embeddings), general ML training (use ml-engineer), video annotation without 3D (use computer-vision-pipeline), or VLM prompt engineering (use prompt-engineer).

curiositech/windags-skills

Implement WCAG 2.2 AA/AAA compliance with automated testing, keyboard navigation, screen reader support, and focus management. Activate on: accessibility audit, WCAG compliance, keyboard navigation, screen reader, aria attributes, axe-core, focus trap. NOT for: design-level accessibility review (use design-accessibility-auditor), color contrast only (use css-in-js-architect).

curiositech/windags-skills

Time-blind friendly planning, executive function support, and daily structure for ADHD brains. Specializes in realistic time estimation, dopamine-aware task design, and building systems that actually work for neurodivergent minds.

curiositech/windags-skills

Designs digital experiences for ADHD brains using neuroscience research and UX principles. Expert in reducing cognitive load, time blindness solutions, dopamine-driven engagement, and compassionate design patterns. Activate on 'ADHD design', 'cognitive load', 'accessibility', 'neurodivergent UX', 'time blindness', 'dopamine-driven', 'executive function'. NOT for general accessibility (WCAG only), neurotypical UX design, or simple UI styling without ADHD context.

curiositech/windags-skills

>- Apply crisis decision-making research to agent routing, uncertainty triage, and coordination failure analysis in time-pressured systems. Use when diagnosing handoff failures, analytical paralysis, or expert judgment under incomplete information. NOT for routine coding, simple CRUD design, or static single-agent tasks with complete information.

curiositech/windags-skills

Extend and modify the admin dashboard, developer portal, and operations console. Use when adding new admin tabs, metrics, monitoring features, or internal tools. Activates for dashboard development, analytics, user management, and internal tooling.

curiositech/windags-skills

Conversation patterns and interaction protocols for multi-agent systems. Covers request/response, pub/sub, blackboard, delegation chains, debate, critique, consensus, fan-out/fan-in, supervisor-worker, and peer negotiation. Deep analysis of AutoGen conversation patterns, CrewAI delegation, LangGraph state passing, and FIPA-ACL performatives. Teaches how to design what agents say to each other and in what order. Activate on: "agent conversation", "agent protocol", "multi-agent debate", "agent delegation", "supervisor worker pattern", "agent voting", "consensus protocol", "fan-out fan-in", "agent negotiation", "blackboard pattern", "agent dialogue", "conversation topology", "agent handoff". NOT for: wire format or serialization (use agent-interchange-formats), orchestration infrastructure (use agentic-infrastructure-2026), single agent behavior (use agentic-patterns).

curiositech/windags-skills

Meta-agent for creating new custom agents, skills, and MCP integrations. Expert in agent design, MCP development, skill architecture, and rapid prototyping. Activate on 'create agent', 'new skill', 'MCP server', 'custom tool', 'agent design'. NOT for using existing agents (invoke them directly), general coding (use language-specific skills), or infrastructure setup (use deployment-engineer).

curiositech/windags-skills

AI-powered calendar management and agent-based scheduling coordination. Covers calendar APIs (Google Calendar, CalDAV/iCal), AI scheduling assistants (Reclaim, Clockwise, Motion, Cal.com), building custom calendar agents with MCP, multi-calendar merging, timezone management, focus block protection, meeting fatigue detection, and agent-to-agent meeting negotiation protocols. Activate on: "calendar agent", "AI scheduling", "calendar coordination", "meeting scheduling", "calendar API", "focus time protection", "calendar optimization", "Google Calendar MCP", "Reclaim", "Clockwise", "Motion", "Cal.com", "smart scheduling", "calendar-aware agent", "timezone scheduling", "agent negotiation meetings". NOT for: manual calendar UI component design (use form-validation-architect), project management scheduling or Gantt charts (use project-management-guru-adhd), general time-tracking or pomodoro apps (use adhd-daily-planner for time-awareness), building the agent itself from scratch (use agent-creator).

curiositech/windags-skills

Build and adopt production AI agent infrastructure in 2026. Covers framework selection (LangGraph, CrewAI, AutoGen, MCP), orchestration patterns, evaluation, observability, memory systems, and tool use. Also covers the SOCIAL dimension: how to sell agent infrastructure internally, change management, measuring ROI, building trust in autonomous systems, and scaling adoption across teams. Activate on: "agent infrastructure", "agent framework comparison", "which agent framework", "sell AI tools internally", "agent adoption", "agent observability", "agent evaluation", "MCP architecture", "agentic mesh", "enterprise AI agents", "AI change management", "agent ROI". NOT for: building specific agents (use ai-engineer), designing agent behavior patterns (use agentic-patterns), prompt tuning (use prompt-engineer).

curiositech/windags-skills

Fundamental patterns for effective agentic behavior. Teaches decomposition, tool orchestration, error recovery, context management, quality self-assessment, and knowing when to stop. Model-agnostic principles that make any agent more effective regardless of domain. Activate on: "how should I structure this agent", "agentic workflow", "agent patterns", "multi-step task", "tool orchestration", "/agentic-patterns", "decompose this", "agent best practices", "chain of actions", "when should the agent stop", "agent loop design". NOT for: creating agent infrastructure (use agent-creator), building DAGs (use windags-architect), specific tool implementation.

curiositech/windags-skills

Cryptographic security for agentic systems — zero-trust agent networking, signed message envelopes (JWS/JWE), capability-based security (ocaps), Merkle tree audit trails, WASM sandboxing, and formal verification. Covers CLI dev tool security, mTLS between agents, permission boundaries (least privilege for AI agents), and supply chain security for skills/plugins. Activate on: "agent security", "zero trust agents", "secure agent communication", "capability-based security", "ocap", "signed messages between agents", "agent audit trail", "sandbox agent execution", "agent permissions", "mTLS agents", "cryptographic verification", "agent supply chain", "OWASP agentic", "prove agent did X", "tamper-proof agent logs". NOT for: application-level SAST scanning (use security-auditor), network firewall rules (use infrastructure), SOC2/HIPAA compliance (organizational), or prompt injection defense (use prompt-engineer).

curiositech/windags-skills

Data structures and serialization formats for agent-to-agent communication. Covers message envelopes, structured output schemas, capability declarations, task handoff payloads, error/retry signaling, and context windows as data structures. Deep comparison of A2A protocol, MCP, OpenAI function calling, and LangChain message types. Teaches when to use rigid schemas vs free-form with validation, typed vs untyped, streaming vs batch. Activate on: "agent message format", "agent communication schema", "agent-to-agent protocol", "A2A protocol", "MCP message format", "structured output for agents", "agent interop", "interchange format", "agent serialization", "task handoff format", "capability declaration". NOT for: what agents say to each other (use agent-conversation-protocols), orchestration topology (use multi-agent-coordination), building agent infrastructure (use agentic-infrastructure-2026).

curiositech/windags-skills

Logic-based agent programming language implementing BDI architecture for practical autonomous agent development

curiositech/windags-skills

>- Design AgentSpeak(L)-style BDI agents with context-guarded plans, selection functions, and intention stacks. Use for interruptible autonomy, agent policy, and multi-agent orchestration in dynamic environments. NOT for simple rule engines, static planners, or centralized workflows.

curiositech/windags-skills

Foundational concurrent computation model where actors communicate exclusively through asynchronous message passing

curiositech/windags-skills

Build production-ready LLM applications, advanced RAG systems, and intelligent agents. Implements vector search, multimodal AI, agent orchestration, and enterprise AI integrations. Use PROACTIVELY for LLM features, chatbots, AI agents, or AI-powered applications.

curiositech/windags-skills

license: Apache-2.0 NOT for unrelated tasks outside this domain.

相关技能