Set up, connect, validate, and repair Claude Desktop and MCP servers using Desktop Commander — inspecting and fixing mcp.json / claude_desktop_config.json, getting MCP servers to connect, and diagnosing local AI tooling. Use when a user wants to install, configure, connect, validate, or fix Claude Desktop, an MCP server, an MCP config file, or a local AI agent/gateway. Trigger on the symptoms that actually bring people in: "Claude can't see my tools", "my MCP server isn't showing up", "tools aren't loading in Claude", "add an MCP server", "my MCP isn't connecting", "fix my Claude Desktop config", "the agent won't respond", "wire up Ollama / LM Studio", "restart Claude Desktop". Also covers local agent gateways like OpenClaw and Hermes.

Parse CI failures, identify failing job/step, classify as flake, regression, env, or config. Use in CI sweeper loops before any fix attempt.

Use when rendering landscape videos with KrillinAI CLI, including original video plus bilingual subtitles or dubbed video plus target-language subtitles.

Use when an agent needs to operate KrillinAI through its CLI, choose the right stage command, inspect CLI outputs, or run subtitle, TTS, landscape render, and portrait render tasks from a YouTube/local video input.

Use this skill ONLY when the user EXPLICITLY requests filing an upstream issue for MoviePilot core, frontend, or an installed plugin, for example "反馈 issue", "提 issue", "报 bug", "给 MP 提 issue", "让上游修一下", "提交错误报告", "提问题", "提需求", "功能请求", or English "file an issue / report a bug / open an upstream issue / feature request". A bare problem report is not enough: diagnose locally first. This skill uses its own scripts under `scripts/`; it does not add or call dedicated Agent tools for collect / prepare / submit.

Use this skill when the user asks to create, scaffold, update, or review a MoviePilot agent skill. This includes adding a new built-in skill under the repository `skills/` directory, editing an existing built-in skill, writing `SKILL.md` frontmatter and workflow instructions, choosing `allowed-tools`, adding helper scripts when needed, and bumping the built-in skill `version` so changes can sync into `config/agent/skills`.

Assists in designing and implementing secure n-tier serverless web applications and microservices on Google Cloud. Use when users need architecture designs, security checklists, Terraform code, or deployment guidance for multi-tier serverless apps, regional data residency / European sovereignty compliance, zero-trust private VPC networking, or Private Service Connect. Don't use for VM, GKE, or non-Google Cloud architectures.

Deploy open models or custom weights from Model Garden to Agent Platform endpoints, check the status of an in-progress deployment operation, or clean up resources by undeploying models and deleting endpoints. Use when asked to actively deploy a model, list the Model Garden CATALOG of available models, check if a specific model is deployable (`gcloud ai model-garden models list-deployment-config`), query deployment cost, troubleshoot deployment errors (like quota limits), or undeploy/clean up endpoints. Also use when copying and deploying a 1P Tuned Model. Don't use for pure listing/discovery questions of the form "is X deployed?", "list my endpoints", or "which regions have models running?" — for those use `agent-platform-endpoint-management`. Don't use for public Vertex AI deployments (use the `vertex-deploy` skill) or for running model evaluations (use the `agent-platform-eval-flywheel` skill).

妙搭（Spark/Miaoda）应用开发与托管：应用创建、本地全栈开发、云端生成迭代、创意设计（UI mockup / 可交互原型 / 线框图 / 落地页 / 仪表盘 / 幻灯片 deck / 视觉探索）、AI相关能力和飞书平台能力或者其他外部能力集成、日志/Trace/监控指标/PV/UV 查询、环境变量管理、应用协作者与协作权限设置、应用角色与成员管理、自动化触发器（定时/记录变更/Webhook/飞书审批）。当用户要开发/新建一个系统·工具·平台·应用，或要本地开发 / 云端开发 / 修改 / 部署 / 发布 / 上线 / 拿可分享链接，或用 HTML 做页面·网站·部署到妙搭，或要设计 / design / mockup / prototype / wireframe / 做 PPT / deck / 视觉探索，或提到妙搭/Spark/Miaoda（应用运行时域名形如 *.aiforce.cloud）、应用数据库、应用文件存储、开放 API Key、可见范围、应用协作者/开发权限、应用角色/角色成员、线上日志、接口请求量、错误量、延迟、访问量、环境变量、给妙搭应用配自动化任务/定时触发/审批通过后自动触发时使用。不负责普通云盘文件上传（lark-drive）、飞书文档编辑（lark-doc）、原生幻灯片创建（lark-slides）。

Drive a PR to a clean review (Greptile 5/5, zero open threads) — ships if needed, keeps it mergeable against staging, re-triggers both Greptile and cubic, fixes real findings, replies to and resolves every thread, and loops until clean

Use when you need exact symbolic math in Python — algebra, calculus, equation solving, symbolic linear algebra, or code generation via lambdify/LaTeX. Prefer NumPy or SciPy when floating-point approximations are sufficient.

Interact with Zotero reference management libraries using the pyzotero Python client. Retrieve, create, update, and delete items, collections, tags, and attachments via the Zotero Web API v3. Use this skill when working with Zotero libraries programmatically, managing bibliographic references, exporting citations, searching library contents, uploading PDF attachments, or building research automation workflows that integrate with Zotero.

Use pydicom to read, inspect, write, transform, and safely preflight local DICOM datasets and pixel data. Applies to DICOM metadata, transfer syntaxes, compression plugins, frames, private elements, JSON, and bounded de-identification review.

Read, validate, and safely export protocols.io data with current official REST/MCP contracts, or create non-executing mutation plans. The bundled client makes bounded official-host GET requests only with explicit --execute. Use only for tasks explicitly targeting protocols.io or an exact protocols.io protocol version.

Convert heterogeneous documents and selected URIs to Markdown with Microsoft MarkItDown for text analysis, search, and LLM/RAG ingestion. Covers safe local conversion, streams, Office/PDF/data formats, batch workflows, plugins, vision OCR, Azure extraction, and the official MCP server.

Plans and audits use of ChicagoHAI HypoGeniC/HypoRefine for LLM-assisted hypothesis generation from labeled text datasets. Use for the `hypogenic` package, its task configs, hypothesis banks, or HypoBench datasets—not for manual hypothesis formulation or scientific validation.

Query the CZ CELLxGENE Census programmatically for versioned public single-cell and spatial transcriptomics data. Use when you need population-scale cell metadata, gene expression slices, Census summary counts, source H5AD URIs/downloads, embeddings, spatial Census data, or reference atlas comparisons across organisms, tissues, diseases, assays, and cell types. For analyzing your own local single-cell data use scanpy, anndata, or scvi-tools.

Test a learner on a single Claude Code tutorial lesson (01-10) with 10 questions, scoring answers and flagging weak spots. Use before, during, or after a lesson. Don't use for whole-tutorial assessment or explaining a topic instead of testing it.

Use when the user asks to review CURRENT uncommitted changes (staged, unstaged, branch diff) before committing, or to synthesize JiT tests against the diff to catch hidden regressions — QA + lead-engineer verdict on completion, quality, architecture, cleanup, plus probes passing on parent and failing on diff. Requires a diff. Do NOT trigger without a diff or on historical commits.

SOLID and DRY structure for an Astro `src/` tree — where files live, size limits, and typed props. Use when deciding where new code belongs, adding or splitting an `.astro` page, layout, or component, or placing a service in `src/lib/` or a type in `src/interfaces/`.