[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M71RDBG0NGX5G6NX24SC77GS":3,"skill-same:01M71RDBG0NGX5G6NX24SC77GS":121},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":25,"publishedAt":45,"linkOnly":46,"markdownContent":47,"markdownBody":48,"frontMatter":49,"refName":55,"commitSha":56,"skillDirectory":57,"forksCount":58,"repositoryHtmlUrl":59,"repositoryHomepageUrl":6,"ownerHtmlUrl":60,"ownerBlogUrl":61,"ownerDescription":6,"updatedAt":62,"install":63,"resources":76},"01M71RDBG0NGX5G6NX24SC77GS","wavebench",null,"Safely diagnose, configure, test, and extend the WaveBench Python measurement bench. Use for WaveBench CLI, run plans, capture packages, reports, TUI, instrument discovery, oscilloscope capture, signal-generator control, programmable-power-supply or digital-multimeter measurements, Core development, and runtime plugin management. Production development in the plugin monorepo uses wavebench-plugin-development; add this skill for Core changes or live operations. Do not use for documentation-led work, general electronics theory, or unrelated VISA\u002FSCPI projects.","Scaxlibur\u002FWaveBench","Scaxlibur","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F51772892?v=4","01M71RD5GE7BHWTFAQ38XVG4XQ","01M71RD5H1R07KNNAP9CQJ7W11",74,"MIT","https:\u002F\u002Fgithub.com\u002FScaxlibur\u002FWaveBench\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Fwavebench\u002FSKILL.md",[17,22],{"categoryNo":18,"categoryName":19,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000008","测试与质量","SCENARIO","使用场景",{"categoryNo":23,"categoryName":24,"taxonomyType":20,"taxonomyTypeDesc":21},"01JBSEEDCAT000000000000023","Skill与智能体开发",[5,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44],"wiring","native","plugin management","cli","unrelated","work","packages","documentation","reports","io","specification","configure","access","host","authority","project","monorepo","power","projects","2026-11-10 13:02:02",false,"---\nname: wavebench\ndescription: >-\n  Safely diagnose, configure, test, and extend the WaveBench Python measurement\n  bench. Use for WaveBench CLI, run plans, capture packages, reports, TUI,\n  instrument discovery, oscilloscope capture, signal-generator control,\n  programmable-power-supply or digital-multimeter measurements, Core development,\n  and runtime plugin management. Production development in the plugin monorepo\n  uses wavebench-plugin-development; add this skill for Core changes or live\n  operations. Do not use for documentation-led work, general electronics theory,\n  or unrelated VISA\u002FSCPI projects.\nlicense: MIT\ncompatibility: >-\n  Codex or a compatible Agent Skills host; Python 3.11+; Linux, WSL, and\n  native Windows are supported. Live instrument operations require the project virtual\n  environment, configured LAN\u002FVISA access, confirmed wiring, and explicit\n  authority for writes.\nmetadata:\n  author: \"WaveBench maintainers\"\n  version: \"2.0.0\"\n  project: \"wavebench\"\n  specification: \"https:\u002F\u002Fagentskills.io\u002Fspecification\"\n---\n\n# WaveBench skill workflow\n\n## Core objective\n\n在不意外改变真实硬件的前提下，完成 WaveBench 的诊断、配置、测量、测试和扩展。优先使用能证明结果的最小操作，先做离线或只读检查，为每次实时写入保留可复核证据。\n\n## Start within the requested scope\n\n1. 用 `git rev-parse --show-toplevel` 定位仓库根目录，并从根目录工作。\n2. 初次进入项目时用 `README.md` 定位入口；涉及依赖或版本时读取 `pyproject.toml`，\n   其余只读取当前任务的实现、契约和相关文档，不预加载全部项目说明。\n   当前 CLI 事实源依次为实现、`--help`、`run schema`、`run template --list` 和\n   `wavebench.example.toml`；技能正文与旧记忆不能覆盖这些事实源。\n3. 执行 `git status --short --branch`，保留无关用户改动；禁止 reset、强制覆盖或隐式清理。\n4. 将任务归类为离线说明\u002F评审、离线代码或配置、实时只读诊断、受控写入或采集。\n5. 在安装依赖、编辑配置或连接硬件前，说明计划、影响范围、预期结果和恢复边界。\n\n解释、诊断和评审默认只读，不自动修复。明确要求实施时，完成已授权修改和相关验证；\n说明计划不是重复确认节点。复合请求按已授权阶段继续，只在关键信息缺失或下一步越界时询问。\n已有实时授权仍需核实当前接线、资源和状态，不能用过去的实验状态代替写前检查。\n\n## Risk classes\n\n| 类别 | 典型操作 | 默认处理 |\n| --- | --- | --- |\n| 离线 | `run schema`、`run template`、`run check`、报告、包检查 | 可直接执行，不连接仪器 |\n| 实时只读 | `doctor`、`idn`、`status`、`profile`、`run verify` | 先说明连接范围；记录有状态查询的副作用 |\n| 实时写入 | setter、output、trigger、fetch、capture、autoscale、TUI、`run plan` | 必须通过写入门禁并在结束后回读 |\n\n`doctor` 会联系配置中的仪器；`run check` 不做仪器 I\u002FO。错误队列、`scope fetch`、通道启用和传输格式设置都可能改变设备状态，不能笼统归为无副作用读取。\n\n## Non-negotiable safety gates\n\n实际连接仪器执行 setter、输出切换、采集、扫频或验收脚本前（离线 fake 测试不适用）：\n\n1. 确认明确的实时写入授权、当前接线和目标资源。\n2. 查询并记录 IDN、相关初始状态、输出状态、保护设置和耦合\u002F负载上下文。\n3. 检查配置的电压、电流、Vpp、频率、超时和重试上限；禁止为通过测试而静默放宽限制。\n4. 按驱动能力执行高阻输入检查；禁止自动加入 `--allow-50ohm` 或等效覆盖项。\n5. 优先使用已通过 `run check` 的计划或既有验收脚本，不把松散命令串当作安全流程。\n6. 明确写前快照、计划恢复范围和不会恢复的设置。\n\n写入期间和结束后：\n\n- 禁止盲目重试写入、触发、采集和输出转换；二进制采集不使用普通读取重试。\n- 写入结果不明确或恢复失败时，保持受影响输出为 OFF，停止后续写入并保存证据。\n- 恢复后重新查询真实仪器，不以进程退出码代替状态确认。\n- 交接中写出产物目录、最终设备状态、未恢复设置和所有跳过或失败的检查。\n\n完整安全语义见 [safety-and-recovery.md](references\u002Fsafety-and-recovery.md)。\n\n## Load only the relevant reference\n\n触发技能后只读取与任务匹配的文件，不预加载整个 `references\u002F`：\n\n| 任务 | 追加读取 |\n| --- | --- |\n| 恢复、状态漂移、停止策略、安全门 | [safety-and-recovery.md](references\u002Fsafety-and-recovery.md) |\n| `run check`、`run verify`、`run plan`、报告、恢复 | [run-plans.md](references\u002Frun-plans.md) |\n| 示波器、通道设置、波形采集 | [scope-and-capture.md](references\u002Fscope-and-capture.md) |\n| 信号发生器、波形、谐波、源状态 | [source-and-harmonics.md](references\u002Fsource-and-harmonics.md) |\n| 可编程电源、数字万用表、保护和测量 | [power-and-dmm.md](references\u002Fpower-and-dmm.md) |\n| 插件发现、检查、安装、升级、回滚 | [plugins.md](references\u002Fplugins.md) |\n| CLI\u002FTUI\u002F报告开发、代码修改、验证和交接 | [development-validation.md](references\u002Fdevelopment-validation.md) |\n| 技能维护或触发回归 | [eval-prompts.md](references\u002Feval-prompts.md) |\n\nReference 只从本入口直接链接，保持一层目录；详细命令和型号边界不得复制回入口。\n\n## Environment and data boundaries\n\n- 要求 Python 3.11+；优先使用项目虚拟环境中的 Python。POSIX 示例使用\n  `.venv\u002Fbin\u002Fpython`，原生 Windows 使用 `.venv\\Scripts\\python.exe`。\n- 原生 Windows 的资源租约依赖 `portalocker[win32]`，默认锁目录为\n  `%LOCALAPPDATA%\\WaveBench\\resource-leases-v1`。Windows 与 WSL 不共享锁域，\n  同一台仪器应固定由一种运行环境访问。\n- `.venv` 不存在或过期时，先说明安装影响；禁止未经授权修改系统 Python。\n- `wavebench.toml` 是本地实验室状态；不要把真实资源地址、序列号或设备标识写入跟踪文件。\n- `data\u002F` 是生成证据；提交采集包、截图、快照或日志前先检查敏感标识。\n- 优先使用 WaveBench CLI、Service 和驱动；裸 SCPI 仅用于明确授权、命令已记录且有安全\u002F恢复方案的驱动探针。\n\n## Discover actual capabilities\n\n当任务依赖当前安装或硬件能力时，不要仅凭型号或 README 推断能力；按需确认已启用的驱动、来源、版本和 capability。纯文档或数值代码修改不需要加载插件：\n\n```bash\n.venv\u002Fbin\u002Fwavebench plugin list --load\n.venv\u002Fbin\u002Fwavebench plugin installed\n.venv\u002Fbin\u002Fwavebench plugin info \u003Cdriver-id> --installed\n.venv\u002Fbin\u002Fwavebench plugin doctor --load\n.venv\u002Fbin\u002Fwavebench capability explain \u003Coperation> --config wavebench.toml\n```\n\n插件 descriptor 的加载会导入可信第三方 Python 代码。能力门拒绝是预期的 fail-closed 结果，不得用裸 SCPI 绕过。\n\n## Standard workflows\n\n按阶段选择预检，不把以下命令作为每个任务的固定清单：\n\n```bash\nuv pip check --python .venv\n.venv\u002Fbin\u002Fwavebench run check --plan plans\u002F\u003Cplan>.toml --config wavebench.toml\n.venv\u002Fbin\u002Fwavebench run verify --plan plans\u002F\u003Cplan>.toml --config wavebench.toml\n```\n\n`uv pip check` 用于依赖变化或环境诊断；`run check` 是离线检查；`run verify` 会连接仪器，只有已授权实时预检时执行。\n已授权执行的真实计划遵循 `run check → run verify → run plan → run report`，并同时检查步骤状态、质量门、期望指标、产物和最终设备状态。仅检查计划的请求到离线检查结果即完成。TUI 界面开发使用 `tui --fake`。\n\n## External research\n\n仓库实现问题优先使用本地事实源。用户要求外部资料，或关键结论需要核实厂商资料、标准或时效信息时，按宿主联网规则检索；遵守用户明确的离线限制，无法核实时说明缺口。优先官方文档，记录来源和日期，不发送本地配置、序列号、网络地址或实验数据。`tavily_hikari` 等搜索 MCP 为可选能力；不可用时说明限制并使用仓库事实源或已能访问的官方页面，不伪造工具调用。\n\n## Code, docs, and handoff\n\n代码改动遵循外科手术式修改：先读实现、契约和聚焦测试，再改最小范围并补测试。公开中文 Markdown 使用项目文档规范，保留代码字面量、路径、URL 和配置键的原样格式。\n\n验证强度按风险匹配，以下是候选命令，不要求每轮全部执行；具体条件见 `development-validation.md`：\n\n```bash\n.venv\u002Fbin\u002Fpython -m pytest -q tests\u002F\u003Cfocused-test>.py\n.venv\u002Fbin\u002Fpython -m pytest -q\n.venv\u002Fbin\u002Fpython -m ruff check .\ngit diff --check\n```\n\n交接先给结论，再列与任务相关的检查结果和限制；产物路径、最终设备状态、未恢复设置只在实际涉及时报告。不得用笼统成功描述掩盖跳过、失败、部分产物或恢复错误。\n","# WaveBench skill workflow\n\n## Core objective\n\n在不意外改变真实硬件的前提下，完成 WaveBench 的诊断、配置、测量、测试和扩展。优先使用能证明结果的最小操作，先做离线或只读检查，为每次实时写入保留可复核证据。\n\n## Start within the requested scope\n\n1. 用 `git rev-parse --show-toplevel` 定位仓库根目录，并从根目录工作。\n2. 初次进入项目时用 `README.md` 定位入口；涉及依赖或版本时读取 `pyproject.toml`，\n   其余只读取当前任务的实现、契约和相关文档，不预加载全部项目说明。\n   当前 CLI 事实源依次为实现、`--help`、`run schema`、`run template --list` 和\n   `wavebench.example.toml`；技能正文与旧记忆不能覆盖这些事实源。\n3. 执行 `git status --short --branch`，保留无关用户改动；禁止 reset、强制覆盖或隐式清理。\n4. 将任务归类为离线说明\u002F评审、离线代码或配置、实时只读诊断、受控写入或采集。\n5. 在安装依赖、编辑配置或连接硬件前，说明计划、影响范围、预期结果和恢复边界。\n\n解释、诊断和评审默认只读，不自动修复。明确要求实施时，完成已授权修改和相关验证；\n说明计划不是重复确认节点。复合请求按已授权阶段继续，只在关键信息缺失或下一步越界时询问。\n已有实时授权仍需核实当前接线、资源和状态，不能用过去的实验状态代替写前检查。\n\n## Risk classes\n\n| 类别 | 典型操作 | 默认处理 |\n| --- | --- | --- |\n| 离线 | `run schema`、`run template`、`run check`、报告、包检查 | 可直接执行，不连接仪器 |\n| 实时只读 | `doctor`、`idn`、`status`、`profile`、`run verify` | 先说明连接范围；记录有状态查询的副作用 |\n| 实时写入 | setter、output、trigger、fetch、capture、autoscale、TUI、`run plan` | 必须通过写入门禁并在结束后回读 |\n\n`doctor` 会联系配置中的仪器；`run check` 不做仪器 I\u002FO。错误队列、`scope fetch`、通道启用和传输格式设置都可能改变设备状态，不能笼统归为无副作用读取。\n\n## Non-negotiable safety gates\n\n实际连接仪器执行 setter、输出切换、采集、扫频或验收脚本前（离线 fake 测试不适用）：\n\n1. 确认明确的实时写入授权、当前接线和目标资源。\n2. 查询并记录 IDN、相关初始状态、输出状态、保护设置和耦合\u002F负载上下文。\n3. 检查配置的电压、电流、Vpp、频率、超时和重试上限；禁止为通过测试而静默放宽限制。\n4. 按驱动能力执行高阻输入检查；禁止自动加入 `--allow-50ohm` 或等效覆盖项。\n5. 优先使用已通过 `run check` 的计划或既有验收脚本，不把松散命令串当作安全流程。\n6. 明确写前快照、计划恢复范围和不会恢复的设置。\n\n写入期间和结束后：\n\n- 禁止盲目重试写入、触发、采集和输出转换；二进制采集不使用普通读取重试。\n- 写入结果不明确或恢复失败时，保持受影响输出为 OFF，停止后续写入并保存证据。\n- 恢复后重新查询真实仪器，不以进程退出码代替状态确认。\n- 交接中写出产物目录、最终设备状态、未恢复设置和所有跳过或失败的检查。\n\n完整安全语义见 [safety-and-recovery.md](references\u002Fsafety-and-recovery.md)。\n\n## Load only the relevant reference\n\n触发技能后只读取与任务匹配的文件，不预加载整个 `references\u002F`：\n\n| 任务 | 追加读取 |\n| --- | --- |\n| 恢复、状态漂移、停止策略、安全门 | [safety-and-recovery.md](references\u002Fsafety-and-recovery.md) |\n| `run check`、`run verify`、`run plan`、报告、恢复 | [run-plans.md](references\u002Frun-plans.md) |\n| 示波器、通道设置、波形采集 | [scope-and-capture.md](references\u002Fscope-and-capture.md) |\n| 信号发生器、波形、谐波、源状态 | [source-and-harmonics.md](references\u002Fsource-and-harmonics.md) |\n| 可编程电源、数字万用表、保护和测量 | [power-and-dmm.md](references\u002Fpower-and-dmm.md) |\n| 插件发现、检查、安装、升级、回滚 | [plugins.md](references\u002Fplugins.md) |\n| CLI\u002FTUI\u002F报告开发、代码修改、验证和交接 | [development-validation.md](references\u002Fdevelopment-validation.md) |\n| 技能维护或触发回归 | [eval-prompts.md](references\u002Feval-prompts.md) |\n\nReference 只从本入口直接链接，保持一层目录；详细命令和型号边界不得复制回入口。\n\n## Environment and data boundaries\n\n- 要求 Python 3.11+；优先使用项目虚拟环境中的 Python。POSIX 示例使用\n  `.venv\u002Fbin\u002Fpython`，原生 Windows 使用 `.venv\\Scripts\\python.exe`。\n- 原生 Windows 的资源租约依赖 `portalocker[win32]`，默认锁目录为\n  `%LOCALAPPDATA%\\WaveBench\\resource-leases-v1`。Windows 与 WSL 不共享锁域，\n  同一台仪器应固定由一种运行环境访问。\n- `.venv` 不存在或过期时，先说明安装影响；禁止未经授权修改系统 Python。\n- `wavebench.toml` 是本地实验室状态；不要把真实资源地址、序列号或设备标识写入跟踪文件。\n- `data\u002F` 是生成证据；提交采集包、截图、快照或日志前先检查敏感标识。\n- 优先使用 WaveBench CLI、Service 和驱动；裸 SCPI 仅用于明确授权、命令已记录且有安全\u002F恢复方案的驱动探针。\n\n## Discover actual capabilities\n\n当任务依赖当前安装或硬件能力时，不要仅凭型号或 README 推断能力；按需确认已启用的驱动、来源、版本和 capability。纯文档或数值代码修改不需要加载插件：\n\n```bash\n.venv\u002Fbin\u002Fwavebench plugin list --load\n.venv\u002Fbin\u002Fwavebench plugin installed\n.venv\u002Fbin\u002Fwavebench plugin info \u003Cdriver-id> --installed\n.venv\u002Fbin\u002Fwavebench plugin doctor --load\n.venv\u002Fbin\u002Fwavebench capability explain \u003Coperation> --config wavebench.toml\n```\n\n插件 descriptor 的加载会导入可信第三方 Python 代码。能力门拒绝是预期的 fail-closed 结果，不得用裸 SCPI 绕过。\n\n## Standard workflows\n\n按阶段选择预检，不把以下命令作为每个任务的固定清单：\n\n```bash\nuv pip check --python .venv\n.venv\u002Fbin\u002Fwavebench run check --plan plans\u002F\u003Cplan>.toml --config wavebench.toml\n.venv\u002Fbin\u002Fwavebench run verify --plan plans\u002F\u003Cplan>.toml --config wavebench.toml\n```\n\n`uv pip check` 用于依赖变化或环境诊断；`run check` 是离线检查；`run verify` 会连接仪器，只有已授权实时预检时执行。\n已授权执行的真实计划遵循 `run check → run verify → run plan → run report`，并同时检查步骤状态、质量门、期望指标、产物和最终设备状态。仅检查计划的请求到离线检查结果即完成。TUI 界面开发使用 `tui --fake`。\n\n## External research\n\n仓库实现问题优先使用本地事实源。用户要求外部资料，或关键结论需要核实厂商资料、标准或时效信息时，按宿主联网规则检索；遵守用户明确的离线限制，无法核实时说明缺口。优先官方文档，记录来源和日期，不发送本地配置、序列号、网络地址或实验数据。`tavily_hikari` 等搜索 MCP 为可选能力；不可用时说明限制并使用仓库事实源或已能访问的官方页面，不伪造工具调用。\n\n## Code, docs, and handoff\n\n代码改动遵循外科手术式修改：先读实现、契约和聚焦测试，再改最小范围并补测试。公开中文 Markdown 使用项目文档规范，保留代码字面量、路径、URL 和配置键的原样格式。\n\n验证强度按风险匹配，以下是候选命令，不要求每轮全部执行；具体条件见 `development-validation.md`：\n\n```bash\n.venv\u002Fbin\u002Fpython -m pytest -q tests\u002F\u003Cfocused-test>.py\n.venv\u002Fbin\u002Fpython -m pytest -q\n.venv\u002Fbin\u002Fpython -m ruff check .\ngit diff --check\n```\n\n交接先给结论，再列与任务相关的检查结果和限制；产物路径、最终设备状态、未恢复设置只在实际涉及时报告。不得用笼统成功描述掩盖跳过、失败、部分产物或恢复错误。\n",{"name":5,"license":14,"metadata":50,"description":7,"compatibility":54},{"author":51,"project":5,"version":52,"specification":53},"WaveBench maintainers","2.0.0","https:\u002F\u002Fagentskills.io\u002Fspecification","Codex or a compatible Agent Skills host; Python 3.11+; Linux, WSL, and native Windows are supported. Live instrument operations require the project virtual environment, configured LAN\u002FVISA access, confirmed wiring, and explicit authority for writes.","master","7677f886e748e6acdb254d069a837e578596d482",".agents\u002Fskills\u002Fwavebench",13,"https:\u002F\u002Fgithub.com\u002FScaxlibur\u002FWaveBench","https:\u002F\u002Fgithub.com\u002FScaxlibur","scaxlibur.github.io","2026-10-26 22:04:14",{"source":8,"skillName":5,"commands":64,"prompt":74,"zipAvailable":75},[65,68,71],{"tool":66,"command":67},"npx","npx skills add Scaxlibur\u002FWaveBench --skill wavebench",{"tool":69,"command":70},"bunx","bunx skills add Scaxlibur\u002FWaveBench --skill wavebench",{"tool":72,"command":73},"pnpm","pnpm dlx skills add Scaxlibur\u002FWaveBench --skill wavebench","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FScaxlibur\u002FWaveBench\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Fwavebench\u002FSKILL.md",true,[77,83,87,91,95,99,103,107,111,115],{"path":78,"resourceType":79,"resourceTypeDesc":80,"sizeBytes":81,"sourceHtmlUrl":82},"agents\u002Fopenai.yaml","REFERENCE","参考文档",281,"https:\u002F\u002Fgithub.com\u002FScaxlibur\u002FWaveBench\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Fwavebench\u002Fagents\u002Fopenai.yaml",{"path":84,"resourceType":79,"resourceTypeDesc":80,"sizeBytes":85,"sourceHtmlUrl":86},"references\u002Fdevelopment-validation.md",4141,"https:\u002F\u002Fgithub.com\u002FScaxlibur\u002FWaveBench\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Fwavebench\u002Freferences\u002Fdevelopment-validation.md",{"path":88,"resourceType":79,"resourceTypeDesc":80,"sizeBytes":89,"sourceHtmlUrl":90},"references\u002Feval-prompts.md",3110,"https:\u002F\u002Fgithub.com\u002FScaxlibur\u002FWaveBench\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Fwavebench\u002Freferences\u002Feval-prompts.md",{"path":92,"resourceType":79,"resourceTypeDesc":80,"sizeBytes":93,"sourceHtmlUrl":94},"references\u002Fplugins.md",3347,"https:\u002F\u002Fgithub.com\u002FScaxlibur\u002FWaveBench\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Fwavebench\u002Freferences\u002Fplugins.md",{"path":96,"resourceType":79,"resourceTypeDesc":80,"sizeBytes":97,"sourceHtmlUrl":98},"references\u002Fpower-and-dmm.md",3465,"https:\u002F\u002Fgithub.com\u002FScaxlibur\u002FWaveBench\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Fwavebench\u002Freferences\u002Fpower-and-dmm.md",{"path":100,"resourceType":79,"resourceTypeDesc":80,"sizeBytes":101,"sourceHtmlUrl":102},"references\u002Frun-plans.md",3920,"https:\u002F\u002Fgithub.com\u002FScaxlibur\u002FWaveBench\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Fwavebench\u002Freferences\u002Frun-plans.md",{"path":104,"resourceType":79,"resourceTypeDesc":80,"sizeBytes":105,"sourceHtmlUrl":106},"references\u002Fsafety-and-recovery.md",4185,"https:\u002F\u002Fgithub.com\u002FScaxlibur\u002FWaveBench\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Fwavebench\u002Freferences\u002Fsafety-and-recovery.md",{"path":108,"resourceType":79,"resourceTypeDesc":80,"sizeBytes":109,"sourceHtmlUrl":110},"references\u002Fscope-and-capture.md",3048,"https:\u002F\u002Fgithub.com\u002FScaxlibur\u002FWaveBench\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Fwavebench\u002Freferences\u002Fscope-and-capture.md",{"path":112,"resourceType":79,"resourceTypeDesc":80,"sizeBytes":113,"sourceHtmlUrl":114},"references\u002Fsource-and-harmonics.md",3440,"https:\u002F\u002Fgithub.com\u002FScaxlibur\u002FWaveBench\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Fwavebench\u002Freferences\u002Fsource-and-harmonics.md",{"path":116,"resourceType":117,"resourceTypeDesc":118,"sizeBytes":119,"sourceHtmlUrl":120},"scripts\u002Fvalidate_skill.py","SCRIPT","脚本",9866,"https:\u002F\u002Fgithub.com\u002FScaxlibur\u002FWaveBench\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Fwavebench\u002Fscripts\u002Fvalidate_skill.py",[122,129],{"skillNo":123,"name":124,"nameZh":6,"description":6,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":125,"sourceHtmlUrl":126,"categories":127,"keywords":128,"publishedAt":45},"01M71RDH7ZV09PTCKZMCNJ063B","WaveBench","https:\u002F\u002Fgithub.com\u002FScaxlibur\u002FWaveBench\u002Fblob\u002Fmaster\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FScaxlibur\u002FWaveBench\u002Fblob\u002Fmaster\u002FSKILL.md",[],[5],{"skillNo":130,"name":131,"nameZh":6,"description":132,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":9,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":133,"categories":134,"keywords":135,"publishedAt":45},"01M71RD87GTJ1XZBE9S379ANK2","wavebench-docs","Develop and maintain the WaveBench repository documentation system. Use only for active documentation development: audits, information-architecture migrations, writing or rewriting project docs, generated references, and documentation-focused diff reviews. Documentation must be a primary requested deliverable; do not use for incidental doc edits during ordinary code changes, merely reading docs, instrument operation, or unrelated Chinese copyediting.","https:\u002F\u002Fgithub.com\u002FScaxlibur\u002FWaveBench\u002Fblob\u002Fmaster\u002F.agents\u002Fskills\u002Fwavebench-docs\u002FSKILL.md",[],[136,137,138,139,140,141,142,30,143,33,144,145,146,147,148,149,150,41,151,152],"docs","wavebench docs","code","doc","reading","repository","active","references","architecture","reviews","information","maintain","writing","generated","only","information-architecture","chinese"]