Explorar skills

Design judgment via design-director taste — reads design intent, assesses whether execution serves it, returns opinionated verdict with clear direction. Scales from one element to a whole app. Always judgment, never fixing. Invoke ONLY via /eye — do NOT trigger on 'review the design' or 'how does it look'.

Opinionated, evolving constraints to guide agents when building interfaces

Jeffrey Emanuel's multi-agent implementation workflow using NTM, Agent Mail, Beads, and BV. The execution phase that follows planning and bead creation. Includes exact prompts used.

A prompt repetition technique for improving LLM accuracy. Achieves significant performance gains in 67% (47/70) of 70 benchmarks. Automatically applied on lightweight models (haiku, flash, mini).

Molecular featurization hub (100+ featurizers) for ML. SMILES to fingerprints (ECFP, MACCS, MAP4), descriptors (RDKit 2D, Mordred), pretrained embeddings (ChemBERTa, GIN, Graphormer), pharmacophores. Scikit-learn compatible with parallelization/caching. For QSAR, virtual screening, similarity, and molecular DL.

Official skill for integrating Firebase AI Logic (Gemini API) into web applications. Covers setup, multimodal inference, structured output, and security.

Generic MCP (Model Context Protocol) server development patterns. Provides reusable architecture and best practices for building MCP servers that expose any domain-specific operations as tools for AI agents. Framework-agnostic implementation supporting async operations, error handling, and enterprise-grade features.

Use when decisions happen async — reviewer(s) look at it over chat, on a PR, on a train, or any time outside the agent session. Triggers: async decision, PR review, pull request, remote reviewer, handoff, away-from-keyboard, decision ballot, architecture call, architecture decision record, ADR, hiring panel, vendor selection, household decision, reconcile, per-reviewer, tick box, ballot.

Pro mode — engineering standard enforcement. Forces senior-engineer behavior: root-cause discipline, layer awareness, current-gen API verification, blast-radius thinking, caller-first API design, no stubs, no sycophantic agreement, no unverified assumptions. Invoke via /prodev or 'prodev' — always run via Skill tool, never skip or summarize.

Destino opcional para cópia de e-mail.

Domine asyncio do Python, programação concorrente e padrões async/await para aplicações de alto desempenho. Utilize ao construir APIs assíncronas, sistemas concorrentes ou aplicações com uso intensivo de I/O que exigem operações não bloqueantes.

Ktor server patterns including routing DSL, plugins, authentication, Koin DI, kotlinx.serialization, WebSockets, and testApplication testing.