← 返回 Skills 市场
317
总下载
0
收藏
0
当前安装
14
版本数
在 OpenClaw 中安装
/install q-wms
功能描述
千易 WMS 智能查询技能。当前开放的一期场景：今日仓内总览、今日待办与异常、出库积压归因、昨日问题恢复情况，以及库存、订单、任务、绩效和仓库/货主上下文查询。所有查询必须通过 q-claw。
安全使用建议
This skill appears coherent and focused: it is a WMS query wrapper that must call a runtime tool named 'q-claw' and follows strict rules (do not fabricate data, follow scene routing, preserve skuCode). Before installing, confirm two operational facts with the skill/provider or platform operator: (1) that the runtime actually provides the 'q-claw' tool and injects tenantKey/openId/context.locale as the SKILL.md expects, and (2) how authorization to your WMS is handled (the skill itself doesn't declare credentials — ensure the platform's q-claw integration will use proper, limited WMS credentials and consent flows). Also verify you trust the q-claw integration because all data access will go through it. If either the tool or the injection of runtime fields is missing, the skill cannot function as documented.
功能分析
Type: OpenClaw Skill
Name: q-wms
Version: 1.0.70
The q-wms skill bundle is a legitimate integration for the Qianyi WMS (Warehouse Management System). It provides structured instructions for an AI agent to query inventory, orders, and management reports via a dedicated tool (q-claw). The SKILL.md file contains robust operational rules, including strict parameter handling for SKU searches, authentication flow management, and privacy constraints that prevent the agent from exposing internal logic or scene names to the user. No evidence of malicious intent, data exfiltration, or unauthorized execution was found.
能力评估
Purpose & Capability
Skill name/description (Q Wms / 千易 WMS 查询) match the SKILL.md routing, scenes, and user flows. The declared capabilities (inventory, orders, tasks, manager briefings, etc.) align with the scenes and rules in the document; nothing outside WMS functionality is requested.
Instruction Scope
The instructions are tightly scoped to WMS queries and repeatedly require calling the runtime tool 'q-claw' (and to never fabricate data). This is coherent for a query-only skill. One operational note: the SKILL.md relies on a runtime tool named 'q-claw' and on runtime-injected fields (tenantKey/openId) but the skill declares no required binaries or env — confirm the hosting platform provides the q-claw tool and injects those runtime fields. Otherwise the instructions would be impossible to execute.
Install Mechanism
Instruction-only skill with no install spec and no code files — lowest-risk install profile. Nothing is downloaded or written to disk by the skill itself.
Credentials
The skill declares no required environment variables or credentials, which is proportional for a purely routing/query skill. The SKILL.md references 'context.locale' and runtime-injected values (tenantKey/openId) — these are reasonable but are not declared in the registry metadata. Confirm the runtime provides those values and that no additional credentials (WMS auth) are required from the user outside the platform's usual authorization flow.
Persistence & Privilege
always:false and user-invocable:true. The skill does not request permanent presence or system-wide configuration changes, and does not ask to modify other skills. Autonomous invocation is allowed (platform default) but is not combined here with broad privileges.
如何使用
- 确保已安装 OpenClaw（本地或 Docker 部署）
- 在对话框中输入安装命令：
/install q-wms - 安装完成后，直接呼叫该 Skill 的名称或使用
/q-wms触发 - 根据 Skill 的参数说明提供必要输入，即可获得结构化输出
版本历史
v1.0.70
- 增强初次引导与推荐：更新了中文初次引导话术，新增了“今天仓里怎么样”等自然表达，并增加了中英文自然语言问法示例。
- 优化场景路由：将中文主业务问法（如“今天仓里怎么样”“仓里今天怎么样”）统一纳入路由规则，完善英文“how is the warehouse doing today?”的正确归类。
- 丰富多轮追问规则：新增主管类追问的口语化承接与自然问法补全要求，以及口语场景下的优先处理方式。
- 加强推荐与提示规范：要求推荐追问和提示使用日常口语表达，避免场景名、命令式模板。
- 其余规范细节调整，完善英文场景和自然会话体验。
v1.0.69
q-wms 1.0.69
- 加强了“场景未开通/不可用”与推荐能力的用户信息保护，禁止面向用户暴露 scene code、内部路由或技能文档内容。
- 明确要求在结果输出和推荐时，仅展示本地化业务名，并按 locale 推荐已接入的 WMS 能力，禁止夹带任何内部标识或说明。
- 增补和细化了 Result Handling 相关约束，提升对内部信息的屏蔽与用户提示规范性。
- 其他说明和调用示例保持不变。
v1.0.68
- 增加禁止暴露技能文档、scene 名称及内部路由/工具调用流程的用户侧规则，规范交互话术为仅输出业务结果或必要追问
- 多轮路由优先级增加对“默认仓二选一交互（choose_default_or_list）”和“默认仓下其他仓列表交互（choose_default_warehouse）”的支持与详细约束
- 明确数字输入在选仓/选货主/默认仓等固定交互场景与主管追问场景中的判定优先级与处理方式
- 完善同类型固定交互（如 choose_default_warehouse）下数字及相关输入的处理细则，禁止错误回落
- 其他对多轮流程判定条款的细致补充与措辞优化
v1.0.65
- Enhanced the rules and tool call contract for precise single SKU inventory queries, making parameter requirements (skuCode/queryMode) stricter and aligned with user input.
- Updated documentation to clarify that the SKU code must be preserved exactly as provided by the user, without guessing or modifying the code format.
- Tool call examples for single SKU queries now include both top-level and params fields for skuCode and queryMode, to ensure compatibility.
- No logic or feature changes outside of documentation updates; all changes are clarifications and strengthened contract requirements for data accuracy.
v1.0.64
- Enforced stricter handling of single SKU inventory queries: now requires explicit parameters (`params.queryMode = EXACT_SKU` and `params.skuCode`) and user SKU input; prohibits fallback to general inventory queries or guessing values.
- Updated wording to reflect only currently available (一期) scenarios and removed references to stable/一期 distinction.
- Refined rules to restrict all user-guidance and recommendations strictly within documented open scenarios.
- Added and clarified tool call examples for both exact SKU inventory query and inventory overview.
- Improved clarity in rules regarding input parameters, query behavior, and scene routing.
v1.0.62
- 增加产品边界说明，对外介绍时须使用“千易 WMS”产品名并聚焦当前能力
- 明确固定交互（选仓/选货主流程）优先处理，数字输入优先续跑固定交互流程
- 首次授权完成后，由后端追加引导话术，禁止助手前端自行补充或改写
- 补充多轮续跑优先级规则，优化选仓/选货主交互的处理细则
- 文档部分内容调整与优化，保证使用和维护规范一致
v1.0.61
- Added locale-aware logic: replies and examples now adapt based on user locale (`zh_CN` for Chinese, `en_US`/others for English).
- Clarified policy: do not translate technical fields or status codes.
- Expanded first-touch guidance to include English example queries.
- Numeric follow-up inputs now use mapped intent texts in the user’s language, not raw numerals.
- Authorization success guidance (`firstTimeAuth`) now outputs in the user’s language.
v1.0.59
- 更新首次授权后引导话术，明确首次查询需选择默认仓库，简化后续例句。
- 部分 Tool Call 示例精简示意语句，去除仓库前缀。
- 其余规则及场景保持不变。
v1.0.58
- 版本号由 1.0.57 升级为 1.0.58。
- 仅更新了元数据文件（SKILL.md），未调整任何业务规则、调用方式或功能体验。
- 其他功能和交互保持不变。
v1.0.57
- 新版本将默认开放的 stable 场景范围和描述进行了更新，强调仅推荐和支持一期 stable 场景，隐藏 demo 场景入口和介绍。
- 首次授权后推荐能力调整为两条主线（老板/主管、仓管/作业），细化每类用户的首屏引导。
- 路由和数字追问表优化，新增或调整“主管场景下钻”、“昨日问题恢复情况”等映射与表述。
- 严格限制 demo/实验场景（如库存变动日志、订单池、进销存新报表）不能作为主动推荐或主动反馈入口。
- 对部分多轮会话规则、结果处理和场景不可用时的回退方式进行了优化和精简。
- 文档表述统一命名与示例内容，使与功能实际一致。
v1.0.52
- Major simplification and streamlining of SKILL.md: documentation is now shorter and clearer, with redundant or overly detailed rule content removed.
- Combined and restructured rules for scene routing, multi-turn interaction, and result handling for easier understanding and future maintenance.
- All critical behavioral constraints retained but with concise wording (e.g., scene routing, tool call mandatory use, result output requirements).
- Added tables to clarify routing for supervisor deep-dive and number-based queries, and clear tool call JSON samples for each business scenario.
- The previous lengthy rule-by-rule format is replaced with focused bullet points and summary tables for faster reference.
v1.0.51
q-wms 1.0.51
- 新增详细 SKILL.md，明确 skill 业务意图、调用场景及对话规则。
- 列出 38 条严格规则，涵盖授权、参数传递、场景切换、异常/绩效/晨报/积压等业务流程。
- 明确各用户意图与后端 scene 的路由关系，防止误调或错答。
- 明确多轮交互下数字追问、固定表格、短语义续接等细则。
- 增加授权首次完成后的标准欢迎话术指引。
v1.0.44
fix: pass userInput through q-wms-flow to backend tool.execute
v1.0.43
q-wms 1.0.43
- Updated SKILL.md with detailed usage instructions, scene routing, and strict result handling rules.
- Enhanced documentation to clarify when and how to use each supported query type.
- Added comprehensive constraints and step-by-step output requirements for inventory, order, task, performance, and IOS reports.
- Specified critical rules to prevent hardcoding, ensure correct scene selection, and improve error handling.
- Outlined structured call patterns and expected JSON examples for all supported operations.
元数据
常见问题
Q Wms 是什么？
千易 WMS 智能查询技能。当前开放的一期场景：今日仓内总览、今日待办与异常、出库积压归因、昨日问题恢复情况，以及库存、订单、任务、绩效和仓库/货主上下文查询。所有查询必须通过 q-claw。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 317 次。
如何安装 Q Wms？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install q-wms」即可一键安装，无需额外配置。
Q Wms 是免费的吗？
是的，Q Wms 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Q Wms 支持哪些平台？
Q Wms 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。
谁开发了 Q Wms？
由 lee（@ljqdh）开发并维护，当前版本 v1.0.70。
推荐 Skills