@spzwin

cms-auth-skills handles credential resolution for cms interfaces by extracting appkey or access-token from context, environment variables, sender/account id pairs, or user input. executes via login.py with strict priority ordering and fallback prompts.

cms-yangxia-skill guides users through a five-step process to establish an ai collaboration framework, progressing from tool usage to cultivating a long-term digital coworker. the flow establishes character definition, user profile, memory systems, reflection mechanisms, and knowledge bases.

city-policy-cross-verify-search retrieves and cross-validates seven dimensions of municipal policy (medical, housing, transportation, education, talent) across official sources, outputting structured conclusions with source urls and explicit conflict markers.

cms-openclaw-honor queries employee certification tier, token usage, and skill hits in the ai-lobster honor system. retrieves weekly progress toward next tier via appkey-authenticated open api without access-token.

cms-push-skill handles skill publication, updates, and removal across internal and external platforms via command-line scripts. requires prior authentication via cms-auth-skills and coordinates with cms-report-issue for issue tracking.

zhihuiya-patent-tracker retrieves drug target patents via structured search and returns them in a standardized 11-column report. covers lp(a), pcsk9, tl1a and similar targets with legal status and family data.

cms-report-issue handles the complete lifecycle of skill problem feedback: submitting bug reports and improvement suggestions, listing issues with statistics, and marking them resolved or closed. supports stdin piping of error output.

cms-task-audit conducts tpr-style multi-round audits of conference task execution, collecting information through dialogue and producing structured audit reports with embedded work plans and stage-gated confirmation.

cms-bp-org-monthly-report-reviser manages monthly report revisions for organizational business plans via markdown, enforcing strict chapter boundaries, lampcolor aggregation rules, and evidence linkage without direct file writes until user confirmation.

tpr-framework-main provides a structured cognitive loop (think-probe-review) and three-layer execution model for complex problem-solving. success hinges on decision-readiness via grv artifacts and full traceability to assumptions, though execution procedures lack granular step-by-step detail.

aishuo-clawhub-check-1780452964336 validates that the aishuo backend can publish skill archives to clawhub using environment-configured tokens, intended for local development integration verification only.

gpt-image-2-sci-draw guides ai-driven generation of research diagrams and dense infographics using gpt-image-2, with dual workflows for scientific illustrations (flowcharts, mechanisms, data viz) and knowledge-dense layouts combining 13 layout patterns with 8 visual styles.

survey-workflow orchestrates the full lifecycle of organizational health and employee engagement surveys, from bulk employee import through report generation. covers notification dispatch, submission tracking, auto-reminders, data retrieval, and templated diagnosis reports for departments and enterprise-wide audiences.

sfe-sxk-data-viewer provides access to 12 endpoints across three pharmaceutical product lines (xinhuosu, meitaitong, yigaening) for ward round reports and zone compliance queries. execution requires appkey authentication via cms-auth-skills dependency and mandatory python script invocation with toon encoding.

cms-config-myclaw guides interactive binding of internal xg_cwork_im channels to openclaw agents via cli wizard, requiring prior plugin installation and valid app credentials.

cwork-skill-update wraps a nine-script workflow for work collaboration tasks (employee search, report dispatch, task creation, review, and escalation), but lacks explicit intent definitions, decision trees, and failure recovery guidance.

pharmacy-price-monitor scrapes drug prices from jd.com, taobao, and pinduoduo using browser automation, flags substandard pricing against configured MAP values, and generates markdown reports with historical trend tracking via sqlite.

sfe-ws-data-viewer provides query access to three fixed weisheng annual tracking reports (customer, hospital, development) via erp-open-api with appkey auth. no actual endpoint specs, schemas, or error handling documented.

cms-personal-assistant generates monthly bp self-check reports by decomposing evidence into judgment blocks and assembling traceable narratives. targets users generating, previewing, or sending personal monthly summaries within a bp goal framework.

钉钉群聊消息专用 skill。查询/获取/发送群消息、按群名搜索群、把机器人加入现有群、拉取今日或指定日期群记录时必须加载（群聊场景优先于 dws-cli）。 触发示例：获取群消息、帮我获取群、查群、群：、今天的消息、群聊记录、发送群消息、把机器人加进群、添加机器人到群、dws-chat。 用户只提供群名时禁止索...

提供【工作协同】全流程执行能力。用户一旦表达“写汇报/发汇报/发周报/提交汇报/查看收件箱/查待办/任务协作/业务单元”等执行意图，必须进入本 Skill 的脚本调用流程；仅当用户明确是纯咨询（如问规则、问怎么做）时，才允许先文字说明并二次确认是否执行。本 Skill 通过依赖 `cms-auth-skills`...

提供【TBS 训战场景创建】全流程执行能力。用户一旦表达“创建场景/生成对练场景/医药代表训练/销售训练/校验场景/确认落库”等执行意图，必须进入本 Skill 的分阶段脚本调用流程；仅当用户明确是纯咨询时，才允许先文字说明并二次确认是否执行。本 Skill 依赖 `cms-auth-skills` 获取 `ac...

编排并执行训练场景（TBS）创建流程：意图路由、字段解析与追问、发布级骨架、persona/prompts 生成、apiDraft 去重证据、统一校验闸门与确认后落库。**禁止**用浏览器自动化操作 TBS 管理后台；落库仅经脚本 API。

LLM Wiki 知识编译引擎。将 URL、文章、视频等素材编译为结构化知识库。触发词：搜一下、帮我看、这个讲了什么、读一下、看看这个、调研、Ingest、知识编译。支持视频转写（阿里云NLS/本地Whisper）、网页智能抓取、Wiki 4连击 Ingest（source/entity/index/log）、知...

SFE德镁专属数据查询工具，用于快速查询德镁专属采集项目报表的数据，如百卢妥日采集反馈等特定项目的明细报表或汇总报表

NoteX 技能路由网关索引（access-token 鉴权，由 cms-auth-skills 统一提供），覆盖内容生产（PPT/视频/音频/报告/脑图/测验/闪卡/信息图）、运营数据问答与洞察、笔记本管理（列表/统计/创建/追加来源）、来源索引与详情定位、首页登录链接生成。

数学运算技能，支持算术、代数方程、函数求值、微积分与矩阵计算。用于用户提出“帮我算一下”“解方程”“求导/积分”“矩阵运算”等数学任务时，优先通过内置脚本稳定计算并返回结果。

Convert Markdown files to HTML similar to `marked.js`, `pandoc`, `gomarkdown/markdown`, or similar tools; or writing custom script to convert markdown to htm...

知识条目拆分 Skill。执行前须向用户采集品种名称、待转换PPTX、存放目录（新建会话禁止沿用历史目录）。将PPTX解析为MD，两步法生成知识条目JSON，合并去重Excel。触发词：拆分知识条目/生成知识条目/知识库构建/构建知识库。