Franklin Silveira Baldo
Community@franklinbaldo · Porto Velho, Rondônia
I craft unique open-source projects and practical AI-powered tools to streamline intricate workflows and tackle real-world challenges.
Agent Skills by Franklin Silveira Baldo
Showing 38 vetted skills indexed across 3 GitHub repositories.
franklinbaldopdf-to-markdown
Converts court process PDFs into structured Markdown files organized by document type.
franklinbaldolicense-enforcement
Audit public evidence of unlicensed skill use and prepare compliance outreach.
franklinbaldovibevoice-asr
Transcribes audio with VibeVoice ASR on Google Colab using BitNet CPU or 7B GPU models.
franklinbaldoverne-orchestration
Orchestrate async Jules coding sessions across repositories via the Verne CLI.
franklinbaldocobogo-design-review
Review web interfaces against Cobogó Brazilian visual grammar using concept-image comparison.
franklinbaldofree-gpu
Execute Python, ML, and inference workloads on Colab or Kaggle GPUs from a local terminal.
franklinbaldopaddleocr
Extract text from PDFs and images using PaddleOCR on local GPU, CPU, or Google Colab.
franklinbaldoconvert-to-pdfa
Convert PDFs to validated PDF/A-1b, PDF/A-2b, or PDF/A-3b archival files using Ghostscript.
franklinbaldosuno-profile
Manages Suno profile bio, genre tags, pinned songs, and playlists via the reverse-engineered Suno write API.
franklinbaldotoe-arena-tiering
Evaluate and tier candidate theories of everything using evidence-based rubrics and clash protocols.
franklinbaldookf-agent-skills
Project Agent Skills repositories into queryable OKF graphs and DuckDB audit views.
franklinbaldolegal-argument-lean
Formalizes and audits Brazilian legal arguments in Lean 4 to expose load-bearing premises.
franklinbaldosoftware-review
Reviews PRs, RFCs, and architectural changes for reproducible defects with calibrated severity.
franklinbaldonotebooklm-processos
Generates grounded NotebookLM question blocks to extract anchored evidence from legal case files.
franklinbaldosuno-curator
Audits and syncs a public Suno music catalog against its blog mirror.
franklinbaldopdf-compression
Compress scanned and image-heavy PDFs by downscaling and re-encoding embedded images.
franklinbaldoopf-finetune
Fine-tune OpenAI Privacy Filter for custom token-classification span tagging of legal text.
franklinbaldocobogo-consumer-synergy
Analyze Cobogó consumer repositories to detect shared design-system needs and coordinate cross-repo issues and PRs.
franklinbaldoloop-engineering
Evolves Agent Skills from real-use postmortems and deduplicated feedback issues.
franklinbaldobrazilian-web-design
Derives accessible web-interface rules from Brazilian visual culture using the Cobogó OKF corpus.
franklinbaldorevisao-minutas
Reviews existing legal drafts to decide whether they can be signed, filed, or issued.
franklinbaldojuris-tjro
Queries TJRO court decisions and full-text rulings through the JURIS search API.
franklinbaldollm-work-via-subagents
Replaces scripted LLM API loops with parallel subagent orchestration for batch tasks.
franklinbaldomeme-image
Generate and embed image memes in blog posts using templates, custom backgrounds, or original visuals.
Frequently Asked Questions About Franklin Silveira Baldo
FAQPage SchemaWhat tasks can I accomplish with Franklin Baldo's skills?▼
You can convert court PDFs to Markdown, compress and archive them as PDF/A, run PaddleOCR on scanned documents, query DataJud/CNJ and TJRO JURIS for case data, anonymize legal texts under LGPD, formalize Brazilian legal arguments in Lean 4, and enforce strict Ruff compliance in code reviews.
Who are these skills designed for?▼
They target Brazilian legal professionals and institutional drafters handling PJe/SEI case files, plus developers maintaining Franklin's open-source projects. Secondary audiences include designers applying Cobogó's Brazilian visual grammar and writers publishing to his blog or Suno catalog.
How do I run these skills in practice?▼
Most skills run via Python with `uv run`, using inline PEP 723 dependencies. Document conversion prefers markitdown or PyMuPDF; GPU-heavy tasks like OPF fine-tuning or VibeVoice transcription route through an authenticated Google Colab or Kaggle CLI session from a local terminal.
Are these skills free and open source?▼
Yes, they are public open-source projects under Franklin Baldo's GitHub account, governed by the Skill Use License 0.1. The license-enforcement skill exists to audit usage beyond that license's permissions and prepare compliance outreach with human approval.
What prerequisites and dependencies are required?▼
Core requirements are Python 3 with uv and network access for court APIs like DataJud and JURIS. PDF/A conversion needs Ghostscript; OCR and anonymization benefit from a CUDA GPU or Colab; locked-down Windows machines use the documented LiteBox adaptation route.