173 skills · sorted by relevance
Create interactive 3D scenes for websites, including product views and scroll-driven animations. Covers choosing Three.js, React Three Fiber, or Spline, preparing lightweight models, showing loading progress, and adapting experiences for mobile devices.
Build text tokenizers that split multilingual text into small units for AI models. Learn how to train SentencePiece on raw text, configure BPE or Unigram splitting methods, and convert text to token IDs and back.
Build custom cognee processing by chaining Python tasks, defining data types, and saving results to graph and vector databases. Use it to enrich existing data, inspect run status, and troubleshoot how tasks pass data.
Review business contracts, including NDAs and vendor agreements, for payment, liability, ownership, and other risks. Get plain-English findings ranked by severity, negotiation suggestions, and an optional DOCX showing proposed deletions and additions.
Create animated Codex pets from character art, text ideas, or brand cues. Generate nine animation states, assemble a transparent sprite sheet, check appearance and movement, and package the results with a pet configuration file.
Turn small game changes into concise specifications with clear reasons, rules, and acceptance criteria. Check existing design documents and earlier specifications, then document tuning adjustments, minor behavior changes, or small features before implementation.
Document a solved and verified problem when its reasoning is missing from code, tests, and existing documentation. Save one durable lesson in the repository so future agents can find it and avoid repeating substantial investigation.
Track machine learning experiments with SwanLab by recording training settings, metrics, and media such as images and audio. Learn to compare runs, inspect training charts, and connect existing training code to local, cloud, or self-hosted dashboards.
Build text tokenizers, which split text into units for AI models, directly from raw multilingual text. Learn custom vocabularies with BPE or Unigram, convert text to pieces or numeric IDs, and decode the results.
Keep context for work that spans multiple sessions without belonging to a specific project phase. The skill manages lightweight persistent threads, letting you create, list, resume, check, and close them as work progresses.
Analyze an existing codebase with parallel agents and produce seven structured documents covering technology, integrations, architecture, conventions, testing, and concerns. Use the results to understand an unfamiliar project or prepare for major changes.
Design or extend DataFusion benchmarks using SQL where practical. Select representative workload variations, keep unrelated processing lightweight, and integrate runs with bench.sh so measurements reflect useful execution paths and remain practical to run.
Manage OpenRig development work from setup to completion using shared specification, progress, and evidence files. Record what was actually checked and pass work to the next owner so handoffs and conversation compaction preserve the context.
Create Word, PDF, and HTML documents from Markdown, choosing pandoc or ReportLab to suit formatting and layout needs. The workflow covers character handling, output checks, and an alternative generation method when PDF conversion fails.
Evaluate an AI product or feature idea by connecting customer problems, business outcomes, testable assumptions, risks, and differentiation. Produce a structured recommendation with lightweight experiments, success measures, and next steps for an investment decision.
Choose a prioritization method that fits your product stage, team, stakeholder needs, and available data. Get a recommended framework, practical setup steps, a scoring example, and an alternative approach with its tradeoffs.
Create formula-based Excel valuation workbooks for public companies using discounted cash flow, a method that converts future cash flows into present value. Include per-share estimates, assumption sensitivities, source records, and checks that flag model limitations.
Add focused runtime logs to a macOS app and verify that expected events appear. Uses Apple’s Logger to trace window, sidebar, menu, and background activity while keeping logs concise and excluding sensitive data.
Reduce GitHub API calls during issue monitoring by scanning basic issue details first, then fetching full descriptions and comments only for selected issues. Filters use assignees, status labels, and title patterns to narrow the list.
Assess factual claims in AI-generated text by separating facts from opinions, choosing suitable independent evidence, and explaining what is supported or uncertain. Use it for a completed fact-check or an optional exercise in evaluating AI answers.
Generate PNG or SVG charts from data for reports, monitoring alerts, and dashboards. Choose from line, bar, pie, candlestick, and heatmap charts, then adjust labels, colors, and annotations using Node.js without a browser or external API calls.
Check whether the MCP connection responds by sending a message and receiving it unchanged, or getting 'pong' when no message is supplied. Useful for CLI health checks, basic SignalR connection tests, and round-trip timing.
Review OpenAPM specification changes through four specialist perspectives covering schemas, registries, dependency resolution, and web architecture. The skill combines findings and mechanical checks into one advisory comment, separating immediate fixes from work deferred to later versions.
Open-ended scientific ideation partner. Use for research gaps, mechanism exploration, interdisciplinary connections, assumptions, possible research directions, and lightweight literature matrix or A+B paper-combination idea mapping. For structured testable hypotheses and validation plans, use hypothesis-generation instead.
The English summary is being prepared. Showing the original description.