Loading...
Loading...
Found 2,800 Skills
Shopee（虾皮）媒体上传 MediaSpace（与 linkfox-shopee-store-auth 同系列），经 /shopee/developerProxy 转发 Shopee Open API MediaSpace 模块全部 6 个接口：init_video_upload、upload_video_part、complete_video_upload、get_video_upload_result、cancel_video_upload、upload_image。当用户提到 Shopee 上传图片、上传视频、media_space、init_video_upload、upload_image、视频分片上传、获取Shopee图片URL 时触发。即使未明确提及"媒体"，只要涉及已授权 Shopee 店铺的图片/视频文件上传，也应触发。
Preflight and diagnose the ten known failure modes for ML training on NVIDIA DGX Spark. Use when a training run on DGX Spark fails to start, OOMs below the 128GB limit, slows down mid-run, or before any multi-hour training job on GB10.
Migrate OpenTelemetry Span Events (AddEvent, RecordException, and language equivalents) to the Logs API following the accepted OTEP 4430 migration plan. Use when migrating instrumentation from span events to log-based events, reviewing code that still uses AddEvent or RecordException, or planning a codebase migration while preserving trace correlation.
Manages spartan/ui, the Angular UI library - adding, composing, fixing, debugging, and styling UI with the Brain (headless primitives) and Helm (styled) layers. Provides project context, component APIs, and usage examples. Applies when working with spartan/ui, @spartan-ng/brain, @spartan-ng/helm, the @spartan-ng/cli generators, or any Angular project with a components.json file. Also triggers for "spartan init", "add a spartan component", or "set up spartan/ui".
Review and finalize multiple pending drafts in a batch session: read context, edit, or discard.
Manage GSD workspaces — create, list, or remove isolated workspace environments
Manage unified memory and thermals during long-running ML jobs on NVIDIA DGX Spark. Use when planning memory headroom for a training run on GB10, when a job OOMs on unified memory, or when monitoring temperature and power during multi-hour training.
Build an interactive Deal Workspace artifact for one or more target accounts - an auto-researched account brief, a buying-group map with deal roles, per-stakeholder dossiers (priorities, talking points, likely objections, a grounded opener) mined from each person's LinkedIn activity, live deal signals, prioritized plays, and a weighted warm-intro chain map that traces every route from your team's LinkedIn connections through real named intermediaries to the decision makers, ranked by how short and strong each route is. Use this whenever someone points at a company or list of companies and wants to know who to sell to and how to get in - especially if they mention or upload LinkedIn Connections.csv exports, or say things like "who do we know at X", "map our warm intros", "who are the decision makers at X", "build a stakeholder map", "find the warmest path into X", "multithread this account", or want account research turned into something shareable with their team. Trigger it even when they only say "research this account" or "help me get a meeting at X" - the warm-path mapping is the whole point.
Manage isolated configurations (tokens, paths) for different contexts.
Concurrent investigation pattern - dispatches multiple AI agents to investigate and fix independent problems simultaneously.
Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies
Use this skill when users need to perform cross-wiki operations on knowledge content across multiple local LLM wikis (yzr-llm-wiki-management system) — from a global workspace perspective: cross-wiki scanning and indexing (INDEX/STATS), cross-wiki Q&A (routing/synthesis/comparison), cross-wiki cross-referencing and duplicate entity governance, workspace-level health checks (LINT), cross-wiki memory (MEMORY/), and workspace format upgrade. Triggers: "Summarize content about X across all my wikis" / "Compare what wiki A and B say about Y" / "Which wiki should I check for this question" / "Scan my workspace" / "Count the number of pages across all wikis" / "Workspace health check" / "Entity X in wiki A also exists in wiki B, add a link" / "Upgrade workspace". Use this skill whenever you need to read, summarize, compare or govern content across multiple wikis — even if the workspace or skill name is not explicitly mentioned. Not applicable: Operations within a single wiki (use yzr-llm-wiki-management); Single llmw command operations such as workspace/wiki metadata configuration, adding/removing wikis, session start/stop (run llmw directly, see llmw --help, no need to load this skill); Cloud collaborative wikis (Notion / Confluence / Outline, etc.).