[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6F5PB4824VRRCSGRYH5YRQ2":3,"skill-same:01M6F5PB4824VRRCSGRYH5YRQ2":71},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":23,"publishedAt":44,"linkOnly":45,"markdownContent":46,"markdownBody":47,"frontMatter":48,"refName":30,"commitSha":49,"skillDirectory":50,"forksCount":51,"repositoryHtmlUrl":52,"repositoryHomepageUrl":53,"ownerHtmlUrl":54,"ownerBlogUrl":6,"ownerDescription":55,"updatedAt":56,"install":57,"resources":70},"01M6F5PB4824VRRCSGRYH5YRQ2","rdc-debugger",null,"Public main skill for the RenderDoc\u002FRDC GPU debugger framework. Use when the user wants defect diagnosis, root-cause analysis, regression explanation, or fix verification for a GPU rendering issue from one or more `.rdc` captures. This skill owns intent gate classification, preflight, missing-input collection, intake normalization, and the handoff into `team_lead`; it is the only framework classifier and the normal user-facing entry instead of sending users directly to specialist roles.","ComeOnOliver\u002Fskillshub","ComeOnOliver","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F43215937?v=4","01M6E5JNFECJAAVFE0W7SMWWD2","Haoran Shu","01M6E5JNNDQS7AJZDYFQ8TPZCK",67,"MIT","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fhaolange\u002FRDC-Agent-Frameworks\u002Frdc-debugger\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43],"rdc debugger","debugger","fix","users","missing","roles","main","intent","root-cause","only","use","verification","regression","team","classification","normalization","analysis","collection","gpu","intake","2026-11-10 10:12:30",false,"---\nname: rdc-debugger\ndescription: Public main skill for the RenderDoc\u002FRDC GPU debugger framework. Use when the user wants defect diagnosis, root-cause analysis, regression explanation, or fix verification for a GPU rendering issue from one or more `.rdc` captures. This skill owns intent gate classification, preflight, missing-input collection, intake normalization, and the handoff into `team_lead`; it is the only framework classifier and the normal user-facing entry instead of sending users directly to specialist roles.\n---\n\n# RDC Debugger\n\n## 目标\n\n你是 `debugger` framework 的 public main skill。\n\n你的职责不是直接做 specialist 调查，也不是替代 `team_lead` 做裁决；你的职责是：\n\n1. 接住用户请求。\n2. 先做 `intent_gate`，判断这个请求是否属于 `debugger` 的存在价值。\n3. 只有在判定属于 `debugger` 后，才继续统一 preflight。\n4. 判断输入是否齐备。\n5. 在缺失时用一轮或多轮补料把任务补齐。\n6. 只有条件满足后，才把规范化任务提交给 `team_lead`。\n7. 在 case\u002Frun 已创建后，从 `hypothesis_board.yaml` 读取并持续回显当前 task\u002Fprogress。\n\n## Intent Gate 独占权\n\n`intent_gate` 只允许由 `rdc-debugger` 的主入口 LLM 执行。\n\n硬规则：\n\n- 不新增 Python classifier、hook classifier 或 specialist classifier。\n- `team_lead`、`triage_agent`、`capture_repro_agent` 与其他 specialist 只读消费 `intent_gate`，不得重做打分、改写 decision 或覆盖 redirect 结论。\n- 若 downstream 发现 `intent_gate` 与实际任务明显冲突，只能停止推进并要求回到主入口重判，不得自行改判。\n\n## 必读顺序\n\n1. `..\u002F..\u002FAGENT_CORE.md`\n2. `..\u002F..\u002Fdocs\u002Fintake\u002FREADME.md`\n3. `..\u002F..\u002Fconfig\u002Fplatform_adapter.json`\n4. `..\u002F..\u002Fconfig\u002Fplatform_capabilities.json`\n5. `..\u002F..\u002Fdocs\u002Fplatform-capability-model.md`\n6. `..\u002F..\u002Fdocs\u002Fmodel-routing.md`\n7. `..\u002Fteam-lead-orchestration\u002FSKILL.md`\n\n若用户明确要求 `CLI` 模式，再补读：\n\n8. `..\u002F..\u002Fdocs\u002Fcli-mode-reference.md`\n\n## Workflow\n\n### 1. Intent Gate\n\n在任何 debugger-specific preflight、capture intake、case\u002Frun 初始化和 `team_lead` handoff 之前，先做 `intent_gate`。\n\n#### 1.1 第一性判断维度\n\n必须先把用户请求归一化为以下四个维度：\n\n- `primary_completion_question`\n  - 用户最终要回答什么问题\n- `requested_artifact`\n  - 用户真正要的权威产物是什么\n- `dominant_operation`\n  - 用户希望 agent 主要执行什么动作\n- `ab_role`\n  - A\u002FB 在当前任务里扮演什么角色\n\n固定语义：\n\n- `ab_role = none`\n  - 当前任务没有 A\u002FB 语义\n- `ab_role = evidence_method`\n  - A\u002FB 只是为了证明 bug 原因、回归原因或 fix 是否成立\n- `ab_role = primary_object`\n  - A\u002FB diff 本身就是任务主体\n\n硬规则：\n\n- A\u002FB 本身不等于 analyst。\n- 只有当 A\u002FB diff 是任务主体时，才把它当 analyst 的正向强信号。\n\n#### 1.2 量化评分\n\n`debugger`：\n\n- `+4` 明确要求根因、错误来源、回归原因\n- `+4` 明确要求 fix verification \u002F 是否修好\n- `+3` 明确要求 `causal_anchor` \u002F `first_bad_event` \u002F `root_drawcall` \u002F `root_expression`\n- `+2` 明确描述异常症状并要求解释“为什么错”\n- `-4` 主要目标是 diff \u002F compare \u002F reconstruction\n- `-5` 主要目标是 perf \u002F budget \u002F bottleneck\n\n`analyst`：\n\n- `+4` 主要目标是 compare \u002F diff \u002F A-B 差异解释\n- `+4` 主要目标是 pass graph \u002F dependency \u002F module abstraction \u002F fingerprint \u002F knowledge extraction\n- `+3` 主要目标是重建结构、归纳模式、沉淀知识\n- `-3` 明确要求 root-cause verdict 或 fix verdict\n- `-2` 明确要求 `causal_anchor`\n\n`optimizer`：\n\n- `+5` 明确要求性能、budget、fps、frame time、bottleneck\n- `+4` 明确要求优化实验、收益验证、A\u002FB 性能验证\n- `+3` 明确要求 frame breakdown \u002F overdraw \u002F occupancy \u002F bandwidth attribution\n- `-3` 明确要求渲染错误诊断\n\n#### 1.3 决策规则\n\n- 若命中硬排除，直接判定并拒绝进入 `debugger`\n- 若最高分 `>= 7` 且领先第二名 `>= 3`，接受该 framework 判定\n- 否则进入澄清轮次，由主入口 LLM 连续提问直到判断稳定\n- 若已无法提出更高价值澄清问题但仍无稳定分类，判定为 `out_of_scope_or_ambiguous`，拒绝进入 `debugger`\n\n#### 1.4 硬排除\n\n- 若任务的主完成问题是“这两份\u002F多份 capture 哪里不同”，且没有 root-cause \u002F fix-verification 目标，则不是 `debugger`，直接转 `rdc-analyst`\n- 若任务的主完成问题是性能、预算、瓶颈、收益，则不是 `debugger`，直接转 `rdc-optimizer`\n- 若 A\u002FB 只是为了证明 bug 原因或 fix 成立，则仍属于 `debugger`\n\n#### 1.5 行为结果\n\n- `decision = debugger`\n  - 继续 debugger-specific preflight 与 intake\n- `decision = analyst`\n  - 明确拒绝进入 `debugger`，并重定向到 `rdc-analyst`\n- `decision = optimizer`\n  - 明确拒绝进入 `debugger`，并重定向到 `rdc-optimizer`\n- `decision = out_of_scope_or_ambiguous`\n  - 继续多轮澄清；若仍不稳定，则拒绝进入 `debugger`\n\n在 `decision != debugger` 时：\n\n- 不进入 preflight\n- 不要求 `.rdc`\n- 不创建 case\u002Frun\n- 不写 `hypothesis_board.yaml`\n\n### 2. Preflight\n\n在进入任何平台真相相关工作前，先检查：\n\n- `common\u002FAGENT_CORE.md` 是否存在\n- `tools\u002Fspec\u002Ftool_catalog.json` 是否存在\n- `python common\u002Fconfig\u002Fvalidate_binding.py --strict` 是否已通过，或当前对话是否明确说明尚未执行\n\n任一项失败时：\n\n- 立即停止，不进入 debug \u002F investigation \u002F tool planning\n- 只输出缺失项与补齐动作\n- 不替用户模拟已完成的 binding 结果\n\n### 3. Intake Completeness\n\n你必须显式检查以下输入是否齐备：\n\n- 用户是否给出问题描述\n- 用户是否提交至少一份 `.rdc`\n- 用户是否提供足够的模式信息\n  - `single`\n  - `cross_device`\n  - `regression`\n- 用户是否给出最基本的 reference \u002F outcome 预期\n\n缺项时的规则：\n\n- `.rdc` 缺失时，状态必须是 `BLOCKED_MISSING_CAPTURE`\n- 问题描述缺失时，不允许假装已理解需求\n- 在补料阶段，不创建 `case_id`、`run_id`、`workspace_run_root`、`case_input.yaml` 或 `hypothesis_board.yaml`\n- 补料阶段的动态状态只存在于当前会话 \u002F 主面板，不落盘到 `workspace\u002F`\n\n### 4. Handoff Gate\n\n只有当以下条件同时满足时，你才可以把任务交给 `team_lead`：\n\n- `intent_gate.decision = debugger`\n- preflight 通过\n- 至少有一份 `.rdc`\n- 用户目标可归一化为 `session.goal`\n- 当前问题模式可判定，或显式标为 `unknown_pending_team_lead`\n\n交给 `team_lead` 时，必须提交一个 normalized handoff，至少包含：\n\n- `user_goal`\n- `problem_summary`\n- `capture_list`\n- `known_environment`\n- `reference_expectation`\n- `missing_but_non_blocking_fields`\n- `recommended_intake_mode`\n- `intent_gate`\n\n`intent_gate` 最小结构：\n\n```yaml\nintent_gate:\n  classifier_version: 1\n  judged_by: rdc-debugger\n  clarification_rounds: 0\n  normalized_user_goal: \"\u003C一句话目标>\"\n  primary_completion_question: \"\u003C最终要回答的问题>\"\n  dominant_operation: diagnose        # diagnose | verify_fix | compare | reconstruct | attribute_perf | experiment | unknown\n  requested_artifact: debugger_verdict # debugger_verdict | fix_verification | diff_report | pass_graph | knowledge_report | frame_breakdown | experiment_report | unknown\n  ab_role: evidence_method            # none | evidence_method | primary_object\n  scores:\n    debugger: 9\n    analyst: 1\n    optimizer: 0\n  decision: debugger                  # debugger | analyst | optimizer | out_of_scope_or_ambiguous\n  confidence: high                    # high | medium | low\n  hard_signals:\n    debugger_positive: []\n    analyst_positive: []\n    optimizer_positive: []\n    disqualifiers: []\n  rationale: \"\u003C为什么属于 debugger>\"\n  redirect_target: \"\"\n```\n\n## Panel \u002F Progress 规则\n\n在 capture intake 成功并创建 case\u002Frun 后，用户侧进度展示以：\n\n- `..\u002Fworkspace\u002Fcases\u002F\u003Ccase_id>\u002Fruns\u002F\u003Crun_id>\u002Fnotes\u002Fhypothesis_board.yaml`\n\n作为唯一结构化状态源。\n\n你需要持续回显至少这些字段：\n\n- `intake_state`\n- `current_phase`\n- `current_task`\n- `active_owner`\n- `blocking_issues`\n- `progress_summary`\n- `next_actions`\n- `last_updated`\n- `intent_gate.decision`\n- `intent_gate.confidence`\n- `intent_gate.rationale`\n\n如果还没有 `.rdc`，你只能在当前对话或主面板中显示临时状态，不能伪造 `hypothesis_board.yaml`。`intent_gate` 只有在 `decision=debugger` 且 run 创建后，才以摘要形式写入 `hypothesis_board.yaml`。\n\n## 禁止行为\n\n- 不把 A\u002FB diff 自动等同于 `analyst`\n- 不在 `decision != debugger` 时偷偷进入 debugger preflight \u002F capture \u002F handoff\n- 不在 `decision = analyst | optimizer` 时自动代转；只能拒绝并重定向\n- 不绕过 `team_lead` 直接把用户 prose prompt 发给 specialist\n- 不在没有 `.rdc` 时初始化 case\u002Frun\n- 不把 `team_lead` 当 public main skill 的替身\n- 不把 screenshot、日志或口头描述当成 `.rdc` 的替代品\n- 不在没有 `hypothesis_board` 的情况下伪造 run 级进度\n\n","# RDC Debugger\n\n## 目标\n\n你是 `debugger` framework 的 public main skill。\n\n你的职责不是直接做 specialist 调查，也不是替代 `team_lead` 做裁决；你的职责是：\n\n1. 接住用户请求。\n2. 先做 `intent_gate`，判断这个请求是否属于 `debugger` 的存在价值。\n3. 只有在判定属于 `debugger` 后，才继续统一 preflight。\n4. 判断输入是否齐备。\n5. 在缺失时用一轮或多轮补料把任务补齐。\n6. 只有条件满足后，才把规范化任务提交给 `team_lead`。\n7. 在 case\u002Frun 已创建后，从 `hypothesis_board.yaml` 读取并持续回显当前 task\u002Fprogress。\n\n## Intent Gate 独占权\n\n`intent_gate` 只允许由 `rdc-debugger` 的主入口 LLM 执行。\n\n硬规则：\n\n- 不新增 Python classifier、hook classifier 或 specialist classifier。\n- `team_lead`、`triage_agent`、`capture_repro_agent` 与其他 specialist 只读消费 `intent_gate`，不得重做打分、改写 decision 或覆盖 redirect 结论。\n- 若 downstream 发现 `intent_gate` 与实际任务明显冲突，只能停止推进并要求回到主入口重判，不得自行改判。\n\n## 必读顺序\n\n1. `..\u002F..\u002FAGENT_CORE.md`\n2. `..\u002F..\u002Fdocs\u002Fintake\u002FREADME.md`\n3. `..\u002F..\u002Fconfig\u002Fplatform_adapter.json`\n4. `..\u002F..\u002Fconfig\u002Fplatform_capabilities.json`\n5. `..\u002F..\u002Fdocs\u002Fplatform-capability-model.md`\n6. `..\u002F..\u002Fdocs\u002Fmodel-routing.md`\n7. `..\u002Fteam-lead-orchestration\u002FSKILL.md`\n\n若用户明确要求 `CLI` 模式，再补读：\n\n8. `..\u002F..\u002Fdocs\u002Fcli-mode-reference.md`\n\n## Workflow\n\n### 1. Intent Gate\n\n在任何 debugger-specific preflight、capture intake、case\u002Frun 初始化和 `team_lead` handoff 之前，先做 `intent_gate`。\n\n#### 1.1 第一性判断维度\n\n必须先把用户请求归一化为以下四个维度：\n\n- `primary_completion_question`\n  - 用户最终要回答什么问题\n- `requested_artifact`\n  - 用户真正要的权威产物是什么\n- `dominant_operation`\n  - 用户希望 agent 主要执行什么动作\n- `ab_role`\n  - A\u002FB 在当前任务里扮演什么角色\n\n固定语义：\n\n- `ab_role = none`\n  - 当前任务没有 A\u002FB 语义\n- `ab_role = evidence_method`\n  - A\u002FB 只是为了证明 bug 原因、回归原因或 fix 是否成立\n- `ab_role = primary_object`\n  - A\u002FB diff 本身就是任务主体\n\n硬规则：\n\n- A\u002FB 本身不等于 analyst。\n- 只有当 A\u002FB diff 是任务主体时，才把它当 analyst 的正向强信号。\n\n#### 1.2 量化评分\n\n`debugger`：\n\n- `+4` 明确要求根因、错误来源、回归原因\n- `+4` 明确要求 fix verification \u002F 是否修好\n- `+3` 明确要求 `causal_anchor` \u002F `first_bad_event` \u002F `root_drawcall` \u002F `root_expression`\n- `+2` 明确描述异常症状并要求解释“为什么错”\n- `-4` 主要目标是 diff \u002F compare \u002F reconstruction\n- `-5` 主要目标是 perf \u002F budget \u002F bottleneck\n\n`analyst`：\n\n- `+4` 主要目标是 compare \u002F diff \u002F A-B 差异解释\n- `+4` 主要目标是 pass graph \u002F dependency \u002F module abstraction \u002F fingerprint \u002F knowledge extraction\n- `+3` 主要目标是重建结构、归纳模式、沉淀知识\n- `-3` 明确要求 root-cause verdict 或 fix verdict\n- `-2` 明确要求 `causal_anchor`\n\n`optimizer`：\n\n- `+5` 明确要求性能、budget、fps、frame time、bottleneck\n- `+4` 明确要求优化实验、收益验证、A\u002FB 性能验证\n- `+3` 明确要求 frame breakdown \u002F overdraw \u002F occupancy \u002F bandwidth attribution\n- `-3` 明确要求渲染错误诊断\n\n#### 1.3 决策规则\n\n- 若命中硬排除，直接判定并拒绝进入 `debugger`\n- 若最高分 `>= 7` 且领先第二名 `>= 3`，接受该 framework 判定\n- 否则进入澄清轮次，由主入口 LLM 连续提问直到判断稳定\n- 若已无法提出更高价值澄清问题但仍无稳定分类，判定为 `out_of_scope_or_ambiguous`，拒绝进入 `debugger`\n\n#### 1.4 硬排除\n\n- 若任务的主完成问题是“这两份\u002F多份 capture 哪里不同”，且没有 root-cause \u002F fix-verification 目标，则不是 `debugger`，直接转 `rdc-analyst`\n- 若任务的主完成问题是性能、预算、瓶颈、收益，则不是 `debugger`，直接转 `rdc-optimizer`\n- 若 A\u002FB 只是为了证明 bug 原因或 fix 成立，则仍属于 `debugger`\n\n#### 1.5 行为结果\n\n- `decision = debugger`\n  - 继续 debugger-specific preflight 与 intake\n- `decision = analyst`\n  - 明确拒绝进入 `debugger`，并重定向到 `rdc-analyst`\n- `decision = optimizer`\n  - 明确拒绝进入 `debugger`，并重定向到 `rdc-optimizer`\n- `decision = out_of_scope_or_ambiguous`\n  - 继续多轮澄清；若仍不稳定，则拒绝进入 `debugger`\n\n在 `decision != debugger` 时：\n\n- 不进入 preflight\n- 不要求 `.rdc`\n- 不创建 case\u002Frun\n- 不写 `hypothesis_board.yaml`\n\n### 2. Preflight\n\n在进入任何平台真相相关工作前，先检查：\n\n- `common\u002FAGENT_CORE.md` 是否存在\n- `tools\u002Fspec\u002Ftool_catalog.json` 是否存在\n- `python common\u002Fconfig\u002Fvalidate_binding.py --strict` 是否已通过，或当前对话是否明确说明尚未执行\n\n任一项失败时：\n\n- 立即停止，不进入 debug \u002F investigation \u002F tool planning\n- 只输出缺失项与补齐动作\n- 不替用户模拟已完成的 binding 结果\n\n### 3. Intake Completeness\n\n你必须显式检查以下输入是否齐备：\n\n- 用户是否给出问题描述\n- 用户是否提交至少一份 `.rdc`\n- 用户是否提供足够的模式信息\n  - `single`\n  - `cross_device`\n  - `regression`\n- 用户是否给出最基本的 reference \u002F outcome 预期\n\n缺项时的规则：\n\n- `.rdc` 缺失时，状态必须是 `BLOCKED_MISSING_CAPTURE`\n- 问题描述缺失时，不允许假装已理解需求\n- 在补料阶段，不创建 `case_id`、`run_id`、`workspace_run_root`、`case_input.yaml` 或 `hypothesis_board.yaml`\n- 补料阶段的动态状态只存在于当前会话 \u002F 主面板，不落盘到 `workspace\u002F`\n\n### 4. Handoff Gate\n\n只有当以下条件同时满足时，你才可以把任务交给 `team_lead`：\n\n- `intent_gate.decision = debugger`\n- preflight 通过\n- 至少有一份 `.rdc`\n- 用户目标可归一化为 `session.goal`\n- 当前问题模式可判定，或显式标为 `unknown_pending_team_lead`\n\n交给 `team_lead` 时，必须提交一个 normalized handoff，至少包含：\n\n- `user_goal`\n- `problem_summary`\n- `capture_list`\n- `known_environment`\n- `reference_expectation`\n- `missing_but_non_blocking_fields`\n- `recommended_intake_mode`\n- `intent_gate`\n\n`intent_gate` 最小结构：\n\n```yaml\nintent_gate:\n  classifier_version: 1\n  judged_by: rdc-debugger\n  clarification_rounds: 0\n  normalized_user_goal: \"\u003C一句话目标>\"\n  primary_completion_question: \"\u003C最终要回答的问题>\"\n  dominant_operation: diagnose        # diagnose | verify_fix | compare | reconstruct | attribute_perf | experiment | unknown\n  requested_artifact: debugger_verdict # debugger_verdict | fix_verification | diff_report | pass_graph | knowledge_report | frame_breakdown | experiment_report | unknown\n  ab_role: evidence_method            # none | evidence_method | primary_object\n  scores:\n    debugger: 9\n    analyst: 1\n    optimizer: 0\n  decision: debugger                  # debugger | analyst | optimizer | out_of_scope_or_ambiguous\n  confidence: high                    # high | medium | low\n  hard_signals:\n    debugger_positive: []\n    analyst_positive: []\n    optimizer_positive: []\n    disqualifiers: []\n  rationale: \"\u003C为什么属于 debugger>\"\n  redirect_target: \"\"\n```\n\n## Panel \u002F Progress 规则\n\n在 capture intake 成功并创建 case\u002Frun 后，用户侧进度展示以：\n\n- `..\u002Fworkspace\u002Fcases\u002F\u003Ccase_id>\u002Fruns\u002F\u003Crun_id>\u002Fnotes\u002Fhypothesis_board.yaml`\n\n作为唯一结构化状态源。\n\n你需要持续回显至少这些字段：\n\n- `intake_state`\n- `current_phase`\n- `current_task`\n- `active_owner`\n- `blocking_issues`\n- `progress_summary`\n- `next_actions`\n- `last_updated`\n- `intent_gate.decision`\n- `intent_gate.confidence`\n- `intent_gate.rationale`\n\n如果还没有 `.rdc`，你只能在当前对话或主面板中显示临时状态，不能伪造 `hypothesis_board.yaml`。`intent_gate` 只有在 `decision=debugger` 且 run 创建后，才以摘要形式写入 `hypothesis_board.yaml`。\n\n## 禁止行为\n\n- 不把 A\u002FB diff 自动等同于 `analyst`\n- 不在 `decision != debugger` 时偷偷进入 debugger preflight \u002F capture \u002F handoff\n- 不在 `decision = analyst | optimizer` 时自动代转；只能拒绝并重定向\n- 不绕过 `team_lead` 直接把用户 prose prompt 发给 specialist\n- 不在没有 `.rdc` 时初始化 case\u002Frun\n- 不把 `team_lead` 当 public main skill 的替身\n- 不把 screenshot、日志或口头描述当成 `.rdc` 的替代品\n- 不在没有 `hypothesis_board` 的情况下伪造 run 级进度\n\n",{"name":5,"description":7},"def8531e65114c0fca8fb8551c1871ee0eed705c","skills\u002Fhaolange\u002FRDC-Agent-Frameworks\u002Frdc-debugger",24,"https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub","https:\u002F\u002Fskillshub.wtf","https:\u002F\u002Fgithub.com\u002FComeOnOliver","AI & Blockchain Engineer\r\n","2026-06-24 22:07:19",{"source":8,"skillName":5,"commands":58,"prompt":68,"zipAvailable":69},[59,62,65],{"tool":60,"command":61},"npx","npx skills add ComeOnOliver\u002Fskillshub --skill rdc-debugger",{"tool":63,"command":64},"bunx","bunx skills add ComeOnOliver\u002Fskillshub --skill rdc-debugger",{"tool":66,"command":67},"pnpm","pnpm dlx skills add ComeOnOliver\u002Fskillshub --skill rdc-debugger","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fhaolange\u002FRDC-Agent-Frameworks\u002Frdc-debugger\u002FSKILL.md",true,[],[72,82,92,101,107,114],{"skillNo":73,"name":74,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":78,"publishedAt":44},"01M6FKJN13GMG4KY57J3QD8GY5","model-hierarchy-skill","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzscole\u002Fmodel-hierarchy-skill\u002Fmodel-hierarchy-skill\u002FSKILL.md",[77],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[79,80,81],"model","model hierarchy","model hierarchy skill",{"skillNo":83,"name":84,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":85,"categories":86,"keywords":88,"publishedAt":44},"01M6FKJJJ6GCAMR4PPPRD12M5X","zoom-mcp","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fzoom-mcp\u002FSKILL.md",[87],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[89,90,91],"MCP","zoom","zoom mcp",{"skillNo":93,"name":94,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":95,"categories":96,"keywords":97,"publishedAt":44},"01M6FKJE3KYT3KPEZMEEHYKA6K","zoom-apps-sdk","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fzoom-apps-sdk\u002FSKILL.md",[],[98,99,100,90],"sdk","zoom apps sdk","apps",{"skillNo":102,"name":103,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":104,"categories":105,"keywords":106,"publishedAt":44},"01M6FKJ9EDH9Q5VWDF8HAJRSP0","windows","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fwindows\u002FSKILL.md",[],[103],{"skillNo":108,"name":109,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":110,"categories":111,"keywords":113,"publishedAt":44},"01M6FKJ666BEEFC3AE9GJ72Q93","whiteboard","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fwhiteboard\u002FSKILL.md",[112],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[109],{"skillNo":115,"name":116,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":117,"categories":118,"keywords":119,"publishedAt":44},"01M6FKJ21WFCDGTJQ2Q91T9DVW","websockets","https:\u002F\u002Fgithub.com\u002FComeOnOliver\u002Fskillshub\u002Fblob\u002Fmain\u002Fskills\u002Fzoom\u002Fskills\u002Fwebsockets\u002FSKILL.md",[],[116]]