GuilhermeBrancalhao
Community@GuilhermeBrancalhao
GuilhermeBrancalhao maintains the AI-ENGINEERING-OS knowledge corpus with 42 volumes, plus language-agnostic code review, performance profiling, architecture decision, and Mermaid diagramming engines.
Agent Skills by GuilhermeBrancalhao
Showing 12 vetted skills indexed across 1 GitHub repositories.
GuilhermeBrancalhaoaieos-cross-reference
Validates dependency graphs and detects semantic contradictions across knowledge volumes.
GuilhermeBrancalhaoaieos-exportar
Generates mkdocs.yml from on-disk content and validates the MkDocs build.
GuilhermeBrancalhaoaieos-auditar
Audits AI-ENGINEERING-OS volumes via a subagent and records dated reports with status updates.
GuilhermeBrancalhaoaieos-novo-volume
Produces a complete AI-ENGINEERING-OS volume with sections, tested examples, and validation gates.
GuilhermeBrancalhaoaieos-status
Reports the status of 42 knowledge volumes by running a read-only Python status module.
GuilhermeBrancalhaorevisar-codigo
Reviews diffs, files, and pull requests, reporting findings ordered by severity with file and line.
GuilhermeBrancalhaootimizar-performance
Diagnoses and fixes performance bottlenecks through measurement-first profiling and quantified validation.
GuilhermeBrancalhaodiagramar
Converts system descriptions and flows into renderable Mermaid diagrams.
GuilhermeBrancalhaomaterializar-ideia
Transforms abstract ideas into running applications with justified stack choices and verified execution.
GuilhermeBrancalhaoarquitetar-sistema
Guides architectural decisions with trade-off analysis, ADR records, and C4 diagrams.
GuilhermeBrancalhaoengine
Orchestrates a persistent phase-based engineering cycle with role agents and risk gates.
GuilhermeBrancalhaoconstruir
Activates the ENGINE phase-driven workflow to build a new project from scratch.
Frequently Asked Questions About GuilhermeBrancalhao
FAQPage SchemaWhat tasks can I accomplish with GuilhermeBrancalhao's skills?▼
You can manage the 42-volume AI-ENGINEERING-OS corpus (create, audit, check status, export to MkDocs, validate cross-references), review code diffs and PRs by severity, profile and fix performance bottlenecks with measured baselines, record architecture decisions as ADRs with C4 diagrams, and turn abstract ideas into running applications.
Who are these skills designed for?▼
Software engineers and technical leads who maintain structured engineering knowledge bases, review code across languages, make architecture trade-off decisions, and need quantified performance work. Java/JVM developers get deeper Spring and JPA criteria via the delegated java-senior engine.
How do the AI-ENGINEERING-OS corpus skills run in practice?▼
They invoke slash commands like /status, /auditar N, /novo-volume, /cross-reference, and /exportar, which execute python -m ferramentas modules for status, validation, and MkDocs generation. The export step runs mkdocs build --strict only when MkDocs is installed, and audits dispatch the auditor-fable subagent.
What prerequisites or dependencies do these skills require?▼
The corpus skills require the ferramentas Python package (status, validar, exportar modules) and optionally MkDocs for site builds. The review, performance, architecture, and diagramming engines are language-agnostic and need no specific runtime, though Java code review delegates to the java-senior engine.
How does the performance optimization skill avoid guesswork?▼
It enforces a Number-First rule: no optimization without measurement. It profiles first, identifies the true bottleneck across algorithm, query, I/O, or network, proves the fix against a reproducible baseline, and delivers a quantified gain rather than speculative changes with maintenance cost.