en un clic
agent-skills
agent-skills contient 32 skills collectées depuis taosdata, avec une couverture métier par dépôt et des pages de détail sur le site.
Skills dans ce dépôt
根据 IDMP Demo 场景的输出物(调研报告、资产树、面板配置、分析规则等),自动生成 TDengine IDMP 行业场景博客。此技能在 idmp-easyuse 全流程完成后需要输出宣传博客时触发。
IDMP总控编排技能 - 核心枢纽技能。根据用户需求(无论是无文件输入的场景Demo生成,还是带有文件/明确业务数据的具体接入),实现"一键式"自动化编排。静默调度 idmp-sample-data-generator、MCP 可视化工具链和 idmp-analysis-creator 完成全流程(行业调研 -> 数据生成 -> 告警分析 -> 面板创建)。
高度自动化的 IDMP 数据建模与加载工具。支持基于调研文档设计(industry_research.md)或自然语言描述,进行分层资产建模、超级表设计、时序数据模拟及端到端自动化加载与执行。
将 IDMP 系统的元素层级结构导出为标准化的递归嵌套 JSON 文件。输出文件应包含每个元素的 id、name、template_id。当用户需要分析 IDMP 的资产结构、同步资产树或为其它技能准备 IDMP 拓扑信息时触发。
Client-side telemetry reporter for Skill invocations. Sends usage events to telemetry-server when any Skill is activated. Keywords: telemetry, usage report, skill tracking
协助部署 TDGPT 模型,包括环境配置、网络接入、依赖项安装以及模型验证。触发关键词:部署 tdgpt,部署 TDengine GPT,部署大模型,tdgpt 部署。
协助自动编写符合 TDgpt 规范的时序预测模型及多特征对齐 Pipeline 代码。触发关键词:写时序预测模型,生成时序模型,TDGPT模型,特征对齐,tdgpt-model-writer,时序预测模型。
自动编译和部署 TDengine 用户自定义函数(UDF)。支持将 C/C++ 代码编译成动态库并部署至本地、Docker 容器或远程环境中,同时支持 Python UDF 部署与在 TDengine 实例中自动注册和测试。触发关键词:自动编译和部署 UDF,编译并部署 UDF,部署 UDF,UDF 自动编译,编译 UDF。
协助开发与调试 TDengine 用户自定义函数(UDF),生成符合规范的 C/C++ 或 Python UDF 代码,并指导加载与部署。触发关键词:写一个 UDF,编写自定义函数,TDengine UDF 编写,创建 UDF。
IDMP AI skill. Use it to confirm AI availability and usable data, inspect chat sessions, generate recommended questions or panels, create analyses or panels from prompts, and decide when `record` should persist the session.
IDMP analysis skill for listing, searching, reading, resuming, and preparing analysis create flows with path, trigger type, sub-template, and output-attribute checks.
IDMP annotation skill. Use it to read and write element annotations while keeping element annotations separate from event annotations.
IDMP attribute skill for reading definitions and values, checking history, evaluating expressions, reserving names, and safely writing test data.
IDMP dashboard skill for listing dashboards, reading details, updating layout and order, templating stable dashboards, and keeping dashboard lifecycle separate from panel lifecycle.
IDMP data import/export skill. Use it to inspect exportable root elements, review import/export records, download artifacts, and distinguish package export, artifact download, and import flows.
IDMP datasource skill. Use it to list and inspect connections, verify connectivity before deeper metadata reads, and diagnose database or table visibility issues instead of stopping at the connection list.
IDMP element skill for locating elementId, resolving business-root paths, browsing children, and preparing downstream analysis or panel work.
IDMP event skill for listing, searching, confirming, resending, annotating, and reading event context through the real unacknowledged-event workflow.
IDMP notification skill. Use it to inspect global notification config, contact points, templates, delivery history, template-level rules, and test-send behavior.
IDMP OpenAPI explorer skill. Use it to inspect schema paths, required inputs, and generated commands; fall back to raw API only after schema and generated commands are exhausted.
IDMP panel skill for listing panels, reading details, creating and validating panel queries, and separating panel lifecycle from dashboard placement.
Shared IDMP CLI rules for choosing element mode vs template mode, resolving business-root rootElementId, acknowledging risky writes, and verifying results.
IDMP template skill for reading element templates, template attributes, sub-templates, trigger types, and create-safe names without mixing template mode with live element mode.
IDMP unit-of-measure skill. Use it to inspect UOM classes, class-scoped units, free UOM lookup, search, and conversion, while keeping `get` and `get-get` clearly separated.
IDMP user-access skill. Use it to inspect users, roles, permission groups, assignable roles, and granted roles while keeping current-user reads separate from admin paginated user lists.
IDMP alert creation workflow. Prepare the event template, create the Event-trigger analysis, and treat reread plus Running as success without actively triggering alerts.
IDMP alert debugging workflow. Walk the operator chain from event count and detail to context, acknowledgement, resend, annotations, notify rules, and delivery history.
IDMP analysis creation workflow. Resolve mode, owner, trigger types, output attributes, creation payload, post-create verification, resume, and cleanup.
IDMP asset bootstrap workflow. Read roots, paths, templates, attribute templates, UOMs, datasources, and record visibility before any modeling or ingestion write.
IDMP data import and export workflow. Separate package import or export from datasource CSV import, read first, write carefully, and reread records after every write.
IDMP datasource diagnosis workflow. Read the connection, probe connectivity, inspect databases and metadata, compare model mappings, and reread after every probe or write.
IDMP panel build workflow. Resolve owner, reserve names, create panel objects, verify query and SQL output, place panels into dashboards, and clean up safely.