{
  "schemaVersion": "1.0",
  "item": {
    "slug": "self-evolving-skill",
    "name": "Self Evolving Skill",
    "source": "tencent",
    "type": "skill",
    "category": "效率提升",
    "sourceUrl": "https://clawhub.ai/whtoo/self-evolving-skill",
    "canonicalUrl": "https://clawhub.ai/whtoo/self-evolving-skill",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/self-evolving-skill",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=self-evolving-skill",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "INSTALLATION.md",
      "README.md",
      "SKILL.md",
      "install.sh",
      "mcporter_adapter.py",
      "openclaw.yaml"
    ],
    "primaryDoc": "SKILL.md",
    "quickSetup": [
      "Download the package from Yavira.",
      "Extract the archive and review SKILL.md first.",
      "Import or place the package into your OpenClaw setup."
    ],
    "agentAssist": {
      "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.",
      "steps": [
        "Download the package from Yavira.",
        "Extract it into a folder your agent can access.",
        "Paste one of the prompts below and point your agent at the extracted folder."
      ],
      "prompts": [
        {
          "label": "New install",
          "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Then review README.md for any prerequisites, environment setup, or post-install checks. Tell me what you changed and call out any manual steps you could not complete."
        },
        {
          "label": "Upgrade existing",
          "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run."
        }
      ]
    },
    "sourceHealth": {
      "source": "tencent",
      "status": "healthy",
      "reason": "direct_download_ok",
      "recommendedAction": "download",
      "checkedAt": "2026-04-30T16:55:25.780Z",
      "expiresAt": "2026-05-07T16:55:25.780Z",
      "httpStatus": 200,
      "finalUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network",
      "contentType": "application/zip",
      "probeMethod": "head",
      "details": {
        "probeUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=network",
        "contentDisposition": "attachment; filename=\"network-1.0.0.zip\"",
        "redirectLocation": null,
        "bodySnippet": null
      },
      "scope": "source",
      "summary": "Source download looks usable.",
      "detail": "Yavira can redirect you to the upstream package for this source.",
      "primaryActionLabel": "Download for OpenClaw",
      "primaryActionHref": "/downloads/self-evolving-skill"
    },
    "validation": {
      "installChecklist": [
        "Use the Yavira download entry.",
        "Review SKILL.md after the package is downloaded.",
        "Confirm the extracted package contains the expected setup assets."
      ],
      "postInstallChecks": [
        "Confirm the extracted package includes the expected docs or setup files.",
        "Validate the skill or prompts are available in your target agent workspace.",
        "Capture any manual follow-up steps the agent could not complete."
      ]
    },
    "downloadPageUrl": "https://openagent3.xyz/downloads/self-evolving-skill",
    "agentPageUrl": "https://openagent3.xyz/skills/self-evolving-skill/agent",
    "manifestUrl": "https://openagent3.xyz/skills/self-evolving-skill/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/self-evolving-skill/agent.md"
  },
  "agentAssist": {
    "summary": "Hand the extracted package to your coding agent with a concrete install brief instead of figuring it out manually.",
    "steps": [
      "Download the package from Yavira.",
      "Extract it into a folder your agent can access.",
      "Paste one of the prompts below and point your agent at the extracted folder."
    ],
    "prompts": [
      {
        "label": "New install",
        "body": "I downloaded a skill package from Yavira. Read SKILL.md from the extracted folder and install it by following the included instructions. Then review README.md for any prerequisites, environment setup, or post-install checks. Tell me what you changed and call out any manual steps you could not complete."
      },
      {
        "label": "Upgrade existing",
        "body": "I downloaded an updated skill package from Yavira. Read SKILL.md from the extracted folder, compare it with my current installation, and upgrade it while preserving any custom configuration unless the package docs explicitly say otherwise. Then review README.md for any prerequisites, environment setup, or post-install checks. Summarize what changed and any follow-up checks I should run."
      }
    ]
  },
  "documentation": {
    "source": "clawhub",
    "primaryDoc": "SKILL.md",
    "sections": [
      {
        "title": "Self-Evolving Skill",
        "body": "元认知自学习系统 - 基于预测编码和价值驱动的Skill自动演化。"
      },
      {
        "title": "功能",
        "body": "ResidualPyramid金字塔分解，量化认知缺口\n-: 残差 自适应反思触发: 基于残差能量自动判断何时需要学习\n经验回放: 缓存已学模式，降低重复触发\n价值门控: 只有提升长期价值才接受变异\n持久化: 经验自动保存/加载"
      },
      {
        "title": "安装",
        "body": "# 技能已安装到 ~/.openclaw/skills/self-evolving-skill\n# 或使用ClawHub\nclawhub install self-evolving-skill"
      },
      {
        "title": "架构",
        "body": "self-evolving-skill/\n├── core/                      # Python核心\n│   ├── residual_pyramid.py     # 残差金字塔（SVD分解）\n│   ├── reflection_trigger.py  # 自适应触发器\n│   ├── experience_replay.py   # 经验回放缓存\n│   ├── skill_engine.py        # 核心引擎+ValueGate\n│   ├── storage.py             # 持久化\n│   └── mcp_server.py          # MCP服务器\n├── src/                       # TypeScript SDK\n│   ├── index.ts               # 主入口\n│   ├── cli.ts                 # CLI\n│   └── mcp-tools.ts           # 工具定义\n├── skills/                    # OpenClaw Skill\n│   └── self-evolving-skill/    # 技能封装\n├── MCP_CONFIG.md              # MCP配置\n└── README.md                   # 文档"
      },
      {
        "title": "MCP工具",
        "body": "工具描述参数skill_create创建Skillname, descriptionskill_execute执行并学习skill_id, context, success, valueskill_analyze分析嵌入embeddingskill_list列出Skills-skill_stats系统统计-skill_save持久化保存skill_idskill_load加载skill_id"
      },
      {
        "title": "CLI",
        "body": "# 列出所有Skill\nopenclaw skill self-evolving-skill list\n\n# 创建Skill\nopenclaw skill self-evolving-skill create --name \"MySkill\"\n\n# 执行\nopenclaw skill self-evolving-skill execute <id> --success\n\n# 分析\nopenclaw skill self-evolving-skill analyze --embedding '[0.1,0.2,...]'\n\n# 统计\nopenclaw skill self-evolving-skill stats"
      },
      {
        "title": "MCP服务器",
        "body": "# 启动MCP服务器\ncd ~/.openclaw/skills/self-evolving-skill\n./run_mcp.sh\n\n# 或使用适配器\npython3 mcporter_adapter.py skill_list '{}'"
      },
      {
        "title": "编程",
        "body": "import { SelfEvolvingSkillEngine } from 'self-evolving-skill';\n\nconst engine = new SelfEvolvingSkillEngine();\nawait engine.init();\n\nconst { skillId } = await engine.createSkill({ name: 'Analyzer' });\nconst stats = await engine.stats();"
      },
      {
        "title": "1. 残差金字塔分解",
        "body": "pyramid = ResidualPyramid(max_layers=5, use_pca=True)\ndecomposition = pyramid.decompose(embedding)\n\n# 输出:\n# - residual_ratio: 残差能量比率\n# - suggested_abstraction: POLICY / SUB_SKILL / PREDICATE\n# - novelty_score: 综合新颖性"
      },
      {
        "title": "2. 三层跃迁规则",
        "body": "覆盖率抽象层级操作>80%POLICY调整策略权重40-80%SUB_SKILL生成子Skill<40%PREDICATE归纳新谓词"
      },
      {
        "title": "3. 自适应阈值",
        "body": "trigger = ReflectionTrigger(\n  min_energy_ratio=0.10,     # 初始阈值\n  value_gain_threshold=0.20, # 触发阈值\n  target_trigger_rate=0.15   # 目标15%触发率\n)"
      },
      {
        "title": "文件位置",
        "body": "路径说明~/.openclaw/skills/self-evolving-skill技能根目录~/.openclaw/mcp_servers/self-evolving-skill.jsonMCP服务器配置~/.openclaw/workspace/self-evolving-skill/storage数据存储"
      },
      {
        "title": "相关文档",
        "body": "README.md - 完整文档\nMCP_CONFIG.md - MCP配置说明\nMEMORY.md - 研究笔记"
      }
    ],
    "body": "Self-Evolving Skill\n\n元认知自学习系统 - 基于预测编码和价值驱动的Skill自动演化。\n\n功能\nResidualPyramid金字塔分解，量化认知缺口 -: 残差 自适应反思触发: 基于残差能量自动判断何时需要学习\n经验回放: 缓存已学模式，降低重复触发\n价值门控: 只有提升长期价值才接受变异\n持久化: 经验自动保存/加载\n安装\n# 技能已安装到 ~/.openclaw/skills/self-evolving-skill\n# 或使用ClawHub\nclawhub install self-evolving-skill\n\n架构\nself-evolving-skill/\n├── core/                      # Python核心\n│   ├── residual_pyramid.py     # 残差金字塔（SVD分解）\n│   ├── reflection_trigger.py  # 自适应触发器\n│   ├── experience_replay.py   # 经验回放缓存\n│   ├── skill_engine.py        # 核心引擎+ValueGate\n│   ├── storage.py             # 持久化\n│   └── mcp_server.py          # MCP服务器\n├── src/                       # TypeScript SDK\n│   ├── index.ts               # 主入口\n│   ├── cli.ts                 # CLI\n│   └── mcp-tools.ts           # 工具定义\n├── skills/                    # OpenClaw Skill\n│   └── self-evolving-skill/    # 技能封装\n├── MCP_CONFIG.md              # MCP配置\n└── README.md                   # 文档\n\nMCP工具\n工具\t描述\t参数\nskill_create\t创建Skill\tname, description\nskill_execute\t执行并学习\tskill_id, context, success, value\nskill_analyze\t分析嵌入\tembedding\nskill_list\t列出Skills\t-\nskill_stats\t系统统计\t-\nskill_save\t持久化保存\tskill_id\nskill_load\t加载\tskill_id\n使用方式\nCLI\n# 列出所有Skill\nopenclaw skill self-evolving-skill list\n\n# 创建Skill\nopenclaw skill self-evolving-skill create --name \"MySkill\"\n\n# 执行\nopenclaw skill self-evolving-skill execute <id> --success\n\n# 分析\nopenclaw skill self-evolving-skill analyze --embedding '[0.1,0.2,...]'\n\n# 统计\nopenclaw skill self-evolving-skill stats\n\nMCP服务器\n# 启动MCP服务器\ncd ~/.openclaw/skills/self-evolving-skill\n./run_mcp.sh\n\n# 或使用适配器\npython3 mcporter_adapter.py skill_list '{}'\n\n编程\nimport { SelfEvolvingSkillEngine } from 'self-evolving-skill';\n\nconst engine = new SelfEvolvingSkillEngine();\nawait engine.init();\n\nconst { skillId } = await engine.createSkill({ name: 'Analyzer' });\nconst stats = await engine.stats();\n\n核心算法\n1. 残差金字塔分解\npyramid = ResidualPyramid(max_layers=5, use_pca=True)\ndecomposition = pyramid.decompose(embedding)\n\n# 输出:\n# - residual_ratio: 残差能量比率\n# - suggested_abstraction: POLICY / SUB_SKILL / PREDICATE\n# - novelty_score: 综合新颖性\n\n2. 三层跃迁规则\n覆盖率\t抽象层级\t操作\n>80%\tPOLICY\t调整策略权重\n40-80%\tSUB_SKILL\t生成子Skill\n<40%\tPREDICATE\t归纳新谓词\n3. 自适应阈值\ntrigger = ReflectionTrigger(\n  min_energy_ratio=0.10,     # 初始阈值\n  value_gain_threshold=0.20, # 触发阈值\n  target_trigger_rate=0.15   # 目标15%触发率\n)\n\n文件位置\n路径\t说明\n~/.openclaw/skills/self-evolving-skill\t技能根目录\n~/.openclaw/mcp_servers/self-evolving-skill.json\tMCP服务器配置\n~/.openclaw/workspace/self-evolving-skill/storage\t数据存储\n相关文档\nREADME.md - 完整文档\nMCP_CONFIG.md - MCP配置说明\nMEMORY.md - 研究笔记"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/whtoo/self-evolving-skill",
    "publisherUrl": "https://clawhub.ai/whtoo/self-evolving-skill",
    "owner": "whtoo",
    "version": "1.0.2",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/self-evolving-skill",
    "downloadUrl": "https://openagent3.xyz/downloads/self-evolving-skill",
    "agentUrl": "https://openagent3.xyz/skills/self-evolving-skill/agent",
    "manifestUrl": "https://openagent3.xyz/skills/self-evolving-skill/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/self-evolving-skill/agent.md"
  }
}