34 skills · sorted by relevance
Develop patent application drafts from an invention description, covering prior art searches, claims that define protection scope, specifications, and examiner-style review. Compile documents for China, the United States, or Europe, with human checkpoints throughout.
Reduce LLM generation delays with speculative decoding, which proposes tokens ahead of verification. This skill compares small draft models, Medusa, and Lookahead Decoding, with implementation examples, tuning guidance, and deployment using vLLM.
Helps speed up LLM text generation by comparing methods that predict and verify multiple tokens at once. Covers draft model selection, Medusa prediction heads, Lookahead Decoding settings, and deployment examples for inference services.
Guides large language model reinforcement learning with miles, including MoE models that route work to specialized components. Covers low-precision training, matching training and inference behavior, and accelerating training response generation with draft models.
Coordinate large research or generation tasks across a model team. An orchestrator plans and checks the work, workers execute subtasks in parallel, and an advisor reviews the plan and draft before the results are combined.
Generates draft requirements from Simulink models. Use when drafting or updating requirement artifacts from a model. Prefers Requirements Toolbox (.slreqx) when available; falls back to structured YAML.
The English summary is being prepared. Showing the original description.
Create and edit FreeCAD models using commands for dimension-controlled sketches, solid parts, assemblies, and technical drawings. Check geometry through previews, measure models, and export files for CAD exchange, 3D printing, or documentation.
Routes listed-company investment research to specialist skills for earnings analysis, valuation, financial model updates, and portfolio risk reviews. Helps organize evidence, assumptions, and investment theses into reports, models, memos, and monitoring trackers.
Edit mathematically, empirically, or technically dense papers for readers outside the author’s specialty, including empirical legal scholarship, law-and-economics models, and legal-technology research. Clarifies terminology, antecedents, study design, and claim–statistic relationships while checking equations, numbers, quotations, and citations against the source. Use to clarify exposition or de-jargonize a technical draft; not to rewrite briefs or doctrinal arguments. Produces an edited draft and audit memo. Full integrity checks and original-format rendering require suitable extraction, code execution, and document tools; otherwise provides proposed edits with verification limits disclosed.
The English summary is being prepared. Showing the original description.
Acelere a inferência de LLMs usando especulative decoding, múltiplas cabeças Medusa e técnicas de lookahead decoding. Use ao otimizar velocidade de inferência (aceleração de 1,5-3,6×), reduzir latência em aplicações em tempo real ou fazer deploy de modelos com recursos computacionais limitados. Cobre modelos draft, atenção em árvore, iteração de Jacobi, geração paralela de tokens e estratégias de deploy em produção.
The English summary is being prepared. Showing the original description.
Role-based multi-model strategy. The model catalog lives in a local configuration file instead of the skill text -- roles (thinker/decision-maker, long-running worker, draft generator, reviewer/acceptance, escalation), reasoning effort as an independent axis, mandatory review for draft models, provider switching for independent reviews, budgets, delegation flow, and cost/quality of small workers.
The English summary is being prepared. Showing the original description.
Plan pricing for APIs, SaaS, and developer tools using customer value and service costs. Compare pricing models, define free tier limits, structure enterprise offers, and assess when and how to raise prices.
Write and edit system prompts, tool instructions, and agent definitions with clear structure and explicit rule strength. It guides tag use, placement of critical constraints, concise wording, and practical examples that show how tools should be used.
Build initial research reports on publicly traded companies, connecting an investment thesis with forecasts, valuation, and risks. Organize supporting sources, distinguish evidence from assumptions, and identify missing information and questions requiring further research.
Generate AI videos using ByteDance Seedance models via Volcengine Ark API. Supports text-to-video, image-to-video (first frame, first+last frame), multimodal reference (images+videos+audio), video editing, video extension, web search enhancement, audio generation, draft mode, offline inference, and continuous video chaining. Use when user wants to generate, create, edit, or extend AI videos from text prompts, images, videos, or audio.
The English summary is being prepared. Showing the original description.
Designs Bayesian clinical trials including Phase I dose-finding (BOIN, CRM, EWOC, mTPI-2), meta-analytic-predictive (MAP) priors with robust mixtures for external data borrowing, EXNEX for basket trials, hierarchical models for safety AE (Berry-Berry), Bayesian platform trials (I-SPY 2, GBM AGILE, REMAP-CAP), and posterior probability stopping rules. Covers FDA Bayesian Devices Guidance (2010), FDA Bayesian Methodology in Drugs Draft (January 2026), BOIN Fit-for-Purpose qualification (December 2021), and Project Optimus dose-optimisation. Use when designing dose-finding studies, platform trials, or sensitivity analyses with informative priors.
The English summary is being prepared. Showing the original description.
Restructure any rough build, research, or legal-drafting request into an IRAC-shaped prompt — Issue, Rule, Analysis, Conclusion — optimized for a frontier model. It's the bar-exam framework, repurposed as prompt engineering. The skill leads with the issue and ends with the conclusion (where models weight attention most), forces you to name your constraints and non-goals, and specifies what "good" looks like before a single token is generated. Use it before any non-trivial build, or whenever a vague ask deserves a precise brief.
The English summary is being prepared. Showing the original description.
Routes any legal task to the right LLM, like OpenRouter but for legal work and grounded in benchmarks instead of brand loyalty. Built from mid-2026 legal evals (legalbenchmarks.ai, Vals AI × Stanford LegalBench across 124 models, Harvey's Legal Agent Benchmark, the Atticus Project's CUAD/MAUD/ACORD) plus translation evidence (WMT25, SwiLTra-Bench, ArabLegalEval). Covers five verticals: contract drafting, info extraction, legal research, contract review, and legal translation (including Arabic/MENA). Each asks up to four questions (cost, speed, accuracy/stakes, privacy/jurisdiction/language), then returns a primary model, a fallback, what to avoid, and what a human must verify. Core principle: capability is not controllability, so every route ends with a verification step. Not legal advice; a lawyer owns the output.
The English summary is being prepared. Showing the original description.
Works out which documents you have to hand over to the other side in a civil case in England & Wales, and builds the formal list. The part it gets right that trips people up is which disclosure regime applies — Practice Direction 57AD (the Disclosure Pilot, now permanent in the Business and Property Courts, with its Models A–E) or standard disclosure under CPR Part 31 everywhere else. It picks the regime, chooses a Model per issue, structures the Disclosure Review Document, and drafts the disclosure certificate for the party to sign personally. Built for litigation juniors, in-house counsel, and small teams without a precedent bank. Use when the user says 'disclosure list', 'PD 57AD', 'Model C', 'extended disclosure', 'List of Documents', 'N265', or needs to work out what must be disclosed.
The English summary is being prepared. Showing the original description.
Split a multi-call LM workflow by cognitive load, not by accuracy: let one strong model make the few reasoning decisions and a cheap model do the many mechanical executions (Aider architect+editor, DSPy optimizer-LM vs task-LM, vLLM speculative draft+target, LangGraph supervisor+worker are the same shape). Use when designing or cost-optimizing a pipeline that calls an LM many times, when deciding which steps need a strong reasoner vs a cheap executor, or when adding an escalation valve for when the cheap tier degrades. Search keywords: reduce LLM cost, cheaper model, lower token cost, model cascade, route to cheap model, strong model plus cheap model, LLM cost optimization.
The English summary is being prepared. Showing the original description.
Designs Bayesian clinical trials including Phase I dose-finding (BOIN, CRM, EWOC, mTPI-2), meta-analytic-predictive (MAP) priors with robust mixtures for external data borrowing, EXNEX for basket trials, hierarchical models for safety AE (Berry-Berry), Bayesian platform trials (I-SPY 2, GBM AGILE, REMAP-CAP), and posterior probability stopping rules. Covers FDA Bayesian Devices Guidance (2010), FDA Bayesian Methodology in Drugs Draft (January 2026), BOIN Fit-for-Purpose qualification (December 2021), and Project Optimus dose-optimisation. Use when designing dose-finding studies, platform trials, or sensitivity analyses with informative priors.
The English summary is being prepared. Showing the original description.
You are an expert in Contentful, the API-first content platform for enterprise teams. You help developers integrate Contentful's Content Delivery API (CDN-backed, read), Content Management API (write), and Content Preview API (draft content) into websites and apps — using typed content models, localization, rich text rendering, image transformations, and webhooks for build triggers.
The English summary is being prepared. Showing the original description.
Writes or reviews a pre-analysis plan before data collection — registry choice (OSF, AEA RCT Registry, AsPredicted, registered reports), PAP structure, models and decision rules locked in advance, analysis code run on simulated data, contingencies for attrition, failed manipulations, and exclusions, deviation records, and timeline. Operationalizes the pre-data-collection side of DA-RT. Use when the user asks to draft or audit a pre-registration or PAP, choose a registry, decide what to lock versus leave exploratory, or handle a later deviation. Hypotheses and estimands come from hypothesis-building, post-hoc reporting from methods-reporting.
The English summary is being prepared. Showing the original description.
Writes or reviews a pre-analysis plan before data collection: registry choice (OSF, AEA, AsPredicted, registered reports), locked models and decision rules, code run on simulated data, contingencies, and deviation records. Use when drafting or auditing a pre-registration or PAP, choosing a registry, or handling a deviation.
The English summary is being prepared. Showing the original description.