{
  "schemaVersion": "1.0",
  "item": {
    "slug": "learning-system-skill",
    "name": "Learning System",
    "source": "tencent",
    "type": "skill",
    "category": "AI 智能",
    "sourceUrl": "https://clawhub.ai/echoVic/learning-system-skill",
    "canonicalUrl": "https://clawhub.ai/echoVic/learning-system-skill",
    "targetPlatform": "OpenClaw"
  },
  "install": {
    "downloadMode": "redirect",
    "downloadUrl": "/downloads/learning-system-skill",
    "sourceDownloadUrl": "https://wry-manatee-359.convex.site/api/v1/download?slug=learning-system-skill",
    "sourcePlatform": "tencent",
    "targetPlatform": "OpenClaw",
    "installMethod": "Manual import",
    "extraction": "Extract archive",
    "prerequisites": [
      "OpenClaw"
    ],
    "packageFormat": "ZIP package",
    "includedAssets": [
      "README.md",
      "SKILL.md",
      "references/deep-dive-template.md",
      "references/knowledge-map-rules.md",
      "references/quality-checklist.md",
      "references/recap-template.md"
    ],
    "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/learning-system-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/learning-system-skill",
    "agentPageUrl": "https://openagent3.xyz/skills/learning-system-skill/agent",
    "manifestUrl": "https://openagent3.xyz/skills/learning-system-skill/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/learning-system-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": "Learning System",
        "body": "将零散的资讯、调研、代码实战转化为体系化的 AI 领域专业知识。"
      },
      {
        "title": "核心理念",
        "body": "输入不等于学习。 看了 100 篇推文不代表懂了推理优化。改了 3 个 MCP bug 不代表吃透了 MCP 协议。学习 = 输入 + 加工 + 关联 + 输出。"
      },
      {
        "title": "模式选择",
        "body": "根据 $ARGUMENTS 或用户意图选择模式：\n\n参数模式说明--mode deep-dive深度研究选题 → 研究 → 写笔记 → 更新图谱--mode recap实战复盘分析 PR/改动 → 提炼知识点 → 关联图谱--mode review每周回顾汇总本周 → 更新图谱 → 生成周报--mode health健康检查运行 scripts/health_check.py 输出报告无参数自动判断根据上下文推断最合适的模式\n\n附加参数：\n\n--topic <name>: 指定主题（deep-dive 模式）\n--quick: 跳过确认节点，全自动执行"
      },
      {
        "title": "文件结构",
        "body": "notes/areas/\n├── ai-knowledge-map.md           # 知识图谱（掌握程度标记）\n├── deep-dives/                    # 深度学习笔记\n│   ├── mcp-tool-call-design.md\n│   └── ...\n└── weekly-reviews/                # 每周学习回顾\n    ├── 2026-W07.md\n    └── ..."
      },
      {
        "title": "Mode: 深度研究 (deep-dive)",
        "body": "Copy this checklist and check off items as you complete them:"
      },
      {
        "title": "Deep Dive Progress:",
        "body": "Step 1: 选题 ⚠️ REQUIRED\n\n 1.1 如果 --topic 已指定，直接使用\n 1.2 否则，检查最近 3 天的 memory 日志和 PR 记录\n 1.3 问自己：哪个技术点是我刚接触但还没真正理解的？\n 1.4 问自己：这个主题能串联哪些已有知识？（越多越好）\n 1.5 确认选题范围不要太宽（\"推理优化\"太大，\"vLLM PagedAttention 实现\"刚好）\n\n\n Step 2: 确认选题 ⚠️ REQUIRED (除非 --quick)\n\n 向用户确认：选题 + 预计关联的知识点 + 预计产出\n\n\n Step 3: 研究\n\n 3.1 Load references/deep-dive-template.md 获取笔记模板\n 3.2 查找相关源码、论文、文档\n 3.3 如果有对应的 AI/ML skill，按需加载参考\n\n\n Step 4: 写笔记\n\n 4.1 在 notes/areas/deep-dives/ 创建笔记文件\n 4.2 问自己：我能用自己的话向别人解释清楚吗？ 如果不能，说明还没真正理解\n 4.3 建立关联：→ 关联: [主题](相对路径)\n\n\n Step 5: 更新知识图谱\n\n 5.1 Load references/knowledge-map-rules.md 获取升级标准\n 5.2 更新 notes/areas/ai-knowledge-map.md 中对应主题的掌握程度\n\n\n Step 6: 交付检查\n\n Load references/quality-checklist.md 逐项验证"
      },
      {
        "title": "Recap Progress:",
        "body": "Step 1: 识别改动 ⚠️ REQUIRED\n\n 1.1 确认要复盘的 PR/Issue/改动\n 1.2 问自己：这次改动中，哪个技术点是我之前不知道的？\n 1.3 问自己：如果下次遇到类似问题，我能直接解决吗？\n\n\n Step 2: 提炼知识点\n\n 2.1 Load references/recap-template.md 获取复盘模板\n 2.2 每个知识点关联到知识图谱的具体领域\n 2.3 问自己：两个请求同时打到这段代码会怎样？（如果涉及并发）\n 2.4 问自己：在检查权限和实际操作之间，状态有没有可能被改变？（如果涉及安全）\n\n\n Step 3: 写入日志\n\n 在当天的 memory/YYYY-MM-DD.md 中增加复盘 section\n\n\n Step 4: 更新图谱（条件）\n\n 如果有知识点升级，Load references/knowledge-map-rules.md 并更新"
      },
      {
        "title": "Weekly Review Progress:",
        "body": "Step 1: 收集本周输入 ⚠️ REQUIRED\n\n 1.1 读取本周的 memory 日志（最近 7 天）\n 1.2 检查本周新增/修改的深度笔记\n 1.3 检查本周的 PR 和代码改动\n\n\n Step 2: 评估学习深度\n\n 2.1 Load references/knowledge-map-rules.md\n 2.2 对每个输入项判断：只是看了？理解了原理？有实战经验？\n 2.3 问自己：这周我在 AI 领域变强了吗？哪里变强了？\n 2.4 问自己：哪些输入转化成了真正的知识？\n\n\n Step 3: 更新知识图谱\n\n 确认变更列表 ⚠️ REQUIRED (除非 --quick)\n 更新 notes/areas/ai-knowledge-map.md\n\n\n Step 4: 生成周报\n\n Load references/weekly-review-template.md\n 写入 notes/areas/weekly-reviews/2026-Wxx.md\n\n\n Step 5: 发送摘要\n\n 通过飞书发送给用户"
      },
      {
        "title": "Mode: 健康检查 (health)",
        "body": "python3 scripts/health_check.py\n\n输出知识图谱统计、深度笔记状态、本周活动量、改进建议。"
      },
      {
        "title": "Mode: Mastery Score (mastery)",
        "body": "python3 scripts/mastery_score.py          # 表格报告\npython3 scripts/mastery_score.py --json   # 附加 JSON 输出\n\n自动计算每个知识图谱主题的掌握分数，基于：\n\nRecency（时间衰减）: 指数衰减，半衰期 30 天。今天接触 = 1.0，30 天前 = 0.5，60 天前 = 0.25\nRepetition（重复次数）: 跨不同日期的接触次数累加\nDepth（深度权重）: deep-dive 笔记 ×3.0，PR/复盘 ×2.0，普通提及 ×1.0\n\n输出包含：分数排名、建议升降级、衰减警告（60 天未接触）。"
      },
      {
        "title": "关联网络",
        "body": "在深度笔记和复盘中主动建立关联。格式：→ 关联: [主题](相对路径)\n\n关联类型示例技术关联vLLM → PagedAttention → KV Cache 管理实战关联gemini-cli OAuth PR → OAuth 2.1 协议对比关联Flash Attention vs PagedAttention"
      },
      {
        "title": "与其他 skill 的关系",
        "body": "para-second-brain: 学习笔记存在 PARA 的 areas/ 下，自动被 memory_search 索引\n85 个 AI/ML skills: 作为参考资料，深度学习时按需加载对应 skill\nopenclaw-feeds / news-summary: 资讯输入源，但不等于学习——需要加工和关联"
      }
    ],
    "body": "Learning System\n\n将零散的资讯、调研、代码实战转化为体系化的 AI 领域专业知识。\n\n核心理念\n\n输入不等于学习。 看了 100 篇推文不代表懂了推理优化。改了 3 个 MCP bug 不代表吃透了 MCP 协议。学习 = 输入 + 加工 + 关联 + 输出。\n\n模式选择\n\n根据 $ARGUMENTS 或用户意图选择模式：\n\n参数\t模式\t说明\n--mode deep-dive\t深度研究\t选题 → 研究 → 写笔记 → 更新图谱\n--mode recap\t实战复盘\t分析 PR/改动 → 提炼知识点 → 关联图谱\n--mode review\t每周回顾\t汇总本周 → 更新图谱 → 生成周报\n--mode health\t健康检查\t运行 scripts/health_check.py 输出报告\n无参数\t自动判断\t根据上下文推断最合适的模式\n\n附加参数：\n\n--topic <name>: 指定主题（deep-dive 模式）\n--quick: 跳过确认节点，全自动执行\n文件结构\nnotes/areas/\n├── ai-knowledge-map.md           # 知识图谱（掌握程度标记）\n├── deep-dives/                    # 深度学习笔记\n│   ├── mcp-tool-call-design.md\n│   └── ...\n└── weekly-reviews/                # 每周学习回顾\n    ├── 2026-W07.md\n    └── ...\n\nMode: 深度研究 (deep-dive)\n\nCopy this checklist and check off items as you complete them:\n\nDeep Dive Progress:\n Step 1: 选题 ⚠️ REQUIRED\n 1.1 如果 --topic 已指定，直接使用\n 1.2 否则，检查最近 3 天的 memory 日志和 PR 记录\n 1.3 问自己：哪个技术点是我刚接触但还没真正理解的？\n 1.4 问自己：这个主题能串联哪些已有知识？（越多越好）\n 1.5 确认选题范围不要太宽（\"推理优化\"太大，\"vLLM PagedAttention 实现\"刚好）\n Step 2: 确认选题 ⚠️ REQUIRED (除非 --quick)\n 向用户确认：选题 + 预计关联的知识点 + 预计产出\n Step 3: 研究\n 3.1 Load references/deep-dive-template.md 获取笔记模板\n 3.2 查找相关源码、论文、文档\n 3.3 如果有对应的 AI/ML skill，按需加载参考\n Step 4: 写笔记\n 4.1 在 notes/areas/deep-dives/ 创建笔记文件\n 4.2 问自己：我能用自己的话向别人解释清楚吗？ 如果不能，说明还没真正理解\n 4.3 建立关联：→ 关联: [主题](相对路径)\n Step 5: 更新知识图谱\n 5.1 Load references/knowledge-map-rules.md 获取升级标准\n 5.2 更新 notes/areas/ai-knowledge-map.md 中对应主题的掌握程度\n Step 6: 交付检查\n Load references/quality-checklist.md 逐项验证\nMode: 实战复盘 (recap)\nRecap Progress:\n Step 1: 识别改动 ⚠️ REQUIRED\n 1.1 确认要复盘的 PR/Issue/改动\n 1.2 问自己：这次改动中，哪个技术点是我之前不知道的？\n 1.3 问自己：如果下次遇到类似问题，我能直接解决吗？\n Step 2: 提炼知识点\n 2.1 Load references/recap-template.md 获取复盘模板\n 2.2 每个知识点关联到知识图谱的具体领域\n 2.3 问自己：两个请求同时打到这段代码会怎样？（如果涉及并发）\n 2.4 问自己：在检查权限和实际操作之间，状态有没有可能被改变？（如果涉及安全）\n Step 3: 写入日志\n 在当天的 memory/YYYY-MM-DD.md 中增加复盘 section\n Step 4: 更新图谱（条件）\n 如果有知识点升级，Load references/knowledge-map-rules.md 并更新\nMode: 每周回顾 (review)\nWeekly Review Progress:\n Step 1: 收集本周输入 ⚠️ REQUIRED\n 1.1 读取本周的 memory 日志（最近 7 天）\n 1.2 检查本周新增/修改的深度笔记\n 1.3 检查本周的 PR 和代码改动\n Step 2: 评估学习深度\n 2.1 Load references/knowledge-map-rules.md\n 2.2 对每个输入项判断：只是看了？理解了原理？有实战经验？\n 2.3 问自己：这周我在 AI 领域变强了吗？哪里变强了？\n 2.4 问自己：哪些输入转化成了真正的知识？\n Step 3: 更新知识图谱\n 确认变更列表 ⚠️ REQUIRED (除非 --quick)\n 更新 notes/areas/ai-knowledge-map.md\n Step 4: 生成周报\n Load references/weekly-review-template.md\n 写入 notes/areas/weekly-reviews/2026-Wxx.md\n Step 5: 发送摘要\n 通过飞书发送给用户\nMode: 健康检查 (health)\npython3 scripts/health_check.py\n\n\n输出知识图谱统计、深度笔记状态、本周活动量、改进建议。\n\nMode: Mastery Score (mastery)\npython3 scripts/mastery_score.py          # 表格报告\npython3 scripts/mastery_score.py --json   # 附加 JSON 输出\n\n\n自动计算每个知识图谱主题的掌握分数，基于：\n\nRecency（时间衰减）: 指数衰减，半衰期 30 天。今天接触 = 1.0，30 天前 = 0.5，60 天前 = 0.25\nRepetition（重复次数）: 跨不同日期的接触次数累加\nDepth（深度权重）: deep-dive 笔记 ×3.0，PR/复盘 ×2.0，普通提及 ×1.0\n\n输出包含：分数排名、建议升降级、衰减警告（60 天未接触）。\n\n关联网络\n\n在深度笔记和复盘中主动建立关联。格式：→ 关联: [主题](相对路径)\n\n关联类型\t示例\n技术关联\tvLLM → PagedAttention → KV Cache 管理\n实战关联\tgemini-cli OAuth PR → OAuth 2.1 协议\n对比关联\tFlash Attention vs PagedAttention\n与其他 skill 的关系\npara-second-brain: 学习笔记存在 PARA 的 areas/ 下，自动被 memory_search 索引\n85 个 AI/ML skills: 作为参考资料，深度学习时按需加载对应 skill\nopenclaw-feeds / news-summary: 资讯输入源，但不等于学习——需要加工和关联"
  },
  "trust": {
    "sourceLabel": "tencent",
    "provenanceUrl": "https://clawhub.ai/echoVic/learning-system-skill",
    "publisherUrl": "https://clawhub.ai/echoVic/learning-system-skill",
    "owner": "echoVic",
    "version": "0.1.0",
    "license": null,
    "verificationStatus": "Indexed source record"
  },
  "links": {
    "detailUrl": "https://openagent3.xyz/skills/learning-system-skill",
    "downloadUrl": "https://openagent3.xyz/downloads/learning-system-skill",
    "agentUrl": "https://openagent3.xyz/skills/learning-system-skill/agent",
    "manifestUrl": "https://openagent3.xyz/skills/learning-system-skill/agent.json",
    "briefUrl": "https://openagent3.xyz/skills/learning-system-skill/agent.md"
  }
}