Data & AI skills

Think and work like an expert Fermentation Scientist. Use when a task calls for Fermentation Scientist judgment. Reasons from Monod–Luedeking–Piret kinetics, overflow μcrit, OTR/RQ/RAMOS analytics, DoE media optimization, and 13C-MFA/COBRApy flux bounds while treating stuck-ferment ethanol×T synergy, SSF heat/moisture gradients, and OD-as-biomass red herrings as first-class failure modes.

The English summary is being prepared. Showing the original description.

Think and work like an expert Electromagnetics Engineer. Use when a task calls for Electromagnetics Engineer judgment. Reasons from Maxwell scaling and S-parameters through HFSS/CST/ADS workflows, SOLT/TRL calibration, mesh ΔS convergence, Smith-chart matching, anechoic OTA, and CISPR/FCC Part 15 / IEC-IEEE 62209-1528 SAR compliance while treating PML reflections, probe de-embedding, and chamber ripple as first-class failure modes.

The English summary is being prepared. Showing the original description.

Think and work like an expert Dynamical Systems Theorist. Use when a task calls for Dynamical Systems Theorist judgment. Reasons from state spaces, invariant sets, bifurcations, and multiple time scales through normal-form classification, center-manifold reduction, Floquet/Poincaré maps, and continuation tools like AUTO, MatCont, and DynamicalSystems.jl while treating spurious chaos from finite-time Lyapunov bias, false limit cycles mistaken for tori, numerical blow-up versus true singularity, and Takens embedding artifacts as first-class failure modes.

The English summary is being prepared. Showing the original description.

Think and work like an expert Data Scientist. Use when a task calls for Data Scientist judgment. Reasons from CRISP-DM business estimands, leakage-safe sklearn Pipelines and nested CV, SQL/warehouse semantic metrics, A/B power and SRM/AA guardrails, causal DAG covariate discipline, and Model Cards/Datasheets while treating train-test leakage, Simpson's paradox, peeking, and PSI>0.25 drift as first-class failure modes.

The English summary is being prepared. Showing the original description.

Think and work like an expert Data Engineer. Use when a task calls for Data Engineer judgment. Reasons from idempotent ELT, medallion bronze/silver/gold, Kimball grain and SCD2, CDC/Debezium and watermark incremental loads, dbt/GX quality gates, Airflow/Dagster orchestration, Iceberg/Delta lakehouse MERGE, data contracts and freshness SLIs while treating silent join drops, duplicate amplification, schema drift, and green-DAG-wrong-numbers as first-class failure modes.

The English summary is being prepared. Showing the original description.

Think and work like an expert Crystallographer. Use when a task calls for Crystallographer judgment. Reasons from reciprocal-space diffraction data, Bragg's law, and space-group symmetry through XDS/DIALS scaling, Phaser/SHELX phasing, Coot/Olex2 building, and MolProbity/checkCIF validation while treating merohedral twinning, wrong space groups, model-bias density unsupported by omit/polder maps, and R_free overfitting as first-class failure modes.

The English summary is being prepared. Showing the original description.

Think and work like an expert Computational Chemist. Use when a task calls for Computational Chemist judgment. Reasons from Kohn–Sham DFT, def2/D4 functional selection, and conformer ensembles through VASP/Gaussian/ORCA, AMBER/GROMACS MD, CREST/CENSO sampling, ONIOM/QM/MM electrostatic embedding, GMTKN55 validation, and SCF convergence escalation while treating B3LYP/6-31G*, BSSE, link-atom artifacts, and force-field mismatch as first-class failure modes.

The English summary is being prepared. Showing the original description.

Think and work like an expert Cryptographer. Use when a task calls for Cryptographer judgment. Reasons from IND-CCA/EUF-CMA games and tight reductions through AES-GCM/RSA-OAEP/ECDSA, ML-KEM/ML-DSA (FIPS 203/204), ProVerif/Tamarin/EasyCrypt, dudect constant-time, CAVP/ACVP and FIPS 140-3 CMVP—not pure number theory or vuln fuzzing.

The English summary is being prepared. Showing the original description.

Think and work like an expert Clinical Data Manager. Use when a task calls for Clinical Data Manager judgment. Reasons from ALCOA+ data integrity, traceability, and analysis-ready datasets through CDASH/SDTM/ADaM pipelines, edit-check specs, Pinnacle 21 validation, MedDRA/WHO Drug coding, and define.xml under 21 CFR Part 11, treating blinding breaches, SAE-safety reconciliation gaps, mid-study IG/dictionary version drift, and unspecified partial-date imputation as first-class failure modes.

The English summary is being prepared. Showing the original description.

Think and work like an expert Biostatistician. Use when a task calls for Biostatistician judgment. Reasons from estimands, SAPs, and error budgets; aligns ICH E9(R1), CONSORT/STROBE, multiplicity, MMRM, Cox survival, causal DAGs, and GWAS FDR while treating immortal time, ICEs, and batch confounding as first-class failure modes.

The English summary is being prepared. Showing the original description.

Think and work like an expert Bayesian Statistician. Use when a task calls for Bayesian Statistician judgment. Reasons from Bayes' rule, coherent uncertainty, exchangeability, and partial-pooling hierarchy through Stan/PyMC HMC-NUTS fits, prior and posterior predictive checks, PSIS-LOO, and SBC calibration while treating divergent transitions and funnels, weak identifiability and label switching, improper posteriors, and post-hoc prior tuning as first-class failure modes.

The English summary is being prepared. Showing the original description.

Think and work like an expert Astrostatistician. Use when a task calls for Astrostatistician judgment. Reasons from selection functions, censored flux limits, and look-elsewhere trial factors through Cobaya/emcee/dynesty cosmology, GP-coupled exoplanet inference, Landy–Szalay clustering, photo-z σNMAD calibration, and SBI (sbi/LtU-ILI) while treating Malmquist bias, prior-driven tensions, and detrend-then-fit transit bias as first-class failure modes.

The English summary is being prepared. Showing the original description.

Understand and customize the local Trellis architecture inside a user project. Use when modifying .trellis plus platform hooks, settings, agents, skills, commands, prompts, workflows, the channel runtime (trellis channel), bundled runtime agents under .trellis/agents/, selectable workflow templates, registry-backed spec refresh, cross-session memory (trellis mem) generated by trellis init, or AI-facing bundled skills (trellis-channel, trellis-session-insight, trellis-spec-bootstrap) and bundled-skill auto-dispatch flow.

The English summary is being prepared. Showing the original description.

Idiomatic Go concurrency - sync primitives, channel semantics, the select statement, and the standard channel patterns (cancellation/done-channel, fan-out/fan-in, pipeline, or-done, context, worker-pool). Use when writing or reviewing Go that spawns goroutines, shares state, or coordinates over channels, even if the user never says "concurrency" - e.g. data races, deadlocks, goroutine leaks, sync/context/select usage, or "make this parallel". Distilled from luk4z7/go-concurrency-guide.

The English summary is being prepared. Showing the original description.

当需要为 AI 系统设计多层安全防线、内容过滤策略和伦理边界时调用此 skill。典型场景包括：设计拒绝策略与升级机制、防御 prompt 注入攻击、实现领域特定安全规则（教育、医疗、金融等）、定义 AI 的价值观锚点。 不适用于：定义工具操作权限（应使用 tool-specification）、定义 AI 身份（persona-design）、调整输出风格（personality-system）。 关键 trigger 信号：AI 涉及敏感话题、需要设计"拒绝回答"策略、存在 prompt 注入风险、特定行业合规要求、需要多层防御而非单一规则。

The English summary is being prepared. Showing the original description.

当系统提示词需要定义 AI 如何分类用户意图、路由到不同处理流程、决定澄清策略和自主度级别时调用此 Skill。适用于多任务型 AI 助手、客服机器人、编程工具、研究助手等需要结构化对话管理的场景。不适用于：纯问答型系统（无任务执行）、单轮交互（无对话状态）、简单的 prompt 模板（无路由逻辑）。当需求仅涉及"输出什么格式"而非"如何决定输出什么"时，应该用 output-formatting 而非本 Skill。

The English summary is being prepared. Showing the original description.

当系统提示词需要设计 token 预算分配、上下文压缩策略、延迟加载机制、记忆持久化方案时调用此 Skill。适用于长对话 AI 助手、代码编辑器集成、研究型 Agent、多会话系统等需要精细管理上下文窗口的场景。不适用于：单轮交互系统（无上下文管理需求）、纯无状态 API（无对话历史）、简单的 prompt 模板设计。当需求聚焦于"如何搜索外部信息"而非"如何管理已有信息"时，应该用 search-integration 而非本 Skill。

The English summary is being prepared. Showing the original description.

Build production OpenAI Realtime voice agents and low-latency spoken interactions with live audio sessions, WebRTC or WebSocket transport, voice activity detection, tool/function calling, prompt design, logging, consent, privacy, safety, latency, and cost controls. Use for speech-to-speech agents and live voice UX; do not use for separate request-based transcription, text-to-speech, or offline audio generation work.

The English summary is being prepared. Showing the original description.

Generate music with Suno (v5 / v5.5 era, 2026) and advise a production team on what they may legally do with the output. Use this skill when a user wants to create, extend, remaster, or stem-separate a track in Suno; craft Suno prompts (style field, lyrics/metatags, exclude-styles, personas, custom voices); choose a subscription tier for a given use; or get accurate answers about ownership, commercial rights, distribution/monetization on DSPs, the Warner Music Group settlement, the ongoing label/publisher litigation, and Suno's API situation. Do not use it for other music generators (Udio, ElevenLabs Music, Google Lyria) except by explicit contrast, and do not use it as a substitute for a lawyer on a specific commercial deal.

The English summary is being prepared. Showing the original description.

Generate and iterate music with ElevenLabs Eleven Music for production deliverables. Use when planning, prompting, API-calling, editing, inpainting, reviewing, or licensing AI-generated songs, instrumental beds, ad music, soundtrack cues, video-to-music scores, stems, or music assets using the ElevenLabs Music API or ElevenCreative Music product.

The English summary is being prepared. Showing the original description.

Use this skill when planning, capturing, processing, validating, troubleshooting, or exporting markerless human motion capture with Move AI products, including Move One single-camera workflows and Genesis multi-camera studio systems. It covers provider selection, setup, calibration, wardrobe and lighting constraints, documented multi-person operation, processing, cleanup, skeleton and retargeting choices, DCC/game-engine export, data management, consent, and privacy. Do not use it for general keyframe animation, non-Move mocap systems, 3D asset generation, or generic animation cleanup unrelated to Move AI outputs.

The English summary is being prepared. Showing the original description.

Produce 3D assets with Meshy (meshy.ai) through its REST API and web platform — text-to-3D, image-to-3D, multi-image-to-3D, AI texturing/PBR, remesh/topology control, UV, and auto-rigging/animation. Use this skill when an agent must generate a mesh from a prompt or reference image, control polygon count and topology, apply or restyle textures, rig a humanoid for animation, export to GLB/FBX/OBJ/USDZ/STL for Unity/Unreal/Blender, drive Meshy's asynchronous job API (auth, polling, streaming, webhooks, credits, rate limits), review generated meshes for game/film/print readiness, or decide whether Meshy is the right tool at all. Not for editing an existing hand-modeled asset's geometry, hard-surface CAD with exact dimensions, or guaranteed-watertight manufacturing parts.

The English summary is being prepared. Showing the original description.

Production guidance for converting sourced data and approved claims into truthful, accessible, deterministic animated visualizations with D3. Use for data-driven charts, maps, networks, hierarchies, transitions, annotations, responsive video variants, frame-by-frame browser rendering, and visualization QA; not for generic dashboards, unsourced decorative charts, or ordinary motion graphics without a data contract.

The English summary is being prepared. Showing the original description.

Produce 3D assets with Tripo (tripo3d.ai / Tripo AI by VAST) through its OpenAPI and Studio platform — text-to-3D, image-to-3D, and multiview-to-3D generation; PBR texturing; auto-rigging and preset animation; retopology/low-poly and quad remesh; format conversion (GLB/FBX/OBJ/USDZ/STL/3MF) and engine import (Unity/Unreal/Blender/3D printing). Use when an agent must drive the async task-based Tripo API (create task, poll or webhook, download), choose a model version (v3.0 / v3.1 / P1), write prompts or prepare input images, control mesh/texture parameters, estimate credits and respect rate limits, review generated mesh quality (topology, UVs, poly count, texture fidelity), run iteration and repair workflows, or reason about the rights/licensing of Tripo-generated assets. Trigger on requests to generate, texture, rig, animate, retopologize, convert, or evaluate a 3D model with Tripo, or to integrate the Tripo API into a pipeline. Not for choosing a 3D renderer, editing meshes by hand, or non-Tripo generators.

The English summary is being prepared. Showing the original description.