anavvanzin
@anavvanzin
Public Skills
fal-ai-media
by anavvanzin
Unified media generation via fal.ai MCP — image, video, and audio. Covers text-to-image (Nano Banana), text/image-to-video (Seedance, Kling, Veo 3), text-to-speech (CSM-1B), and video-to-audio (ThinkSound). Use when the user wants to generate images, videos, or audio with AI.
frontend-patterns
by anavvanzin
Frontend development patterns for React, Next.js, state management, performance optimization, and UI best practices.
video-editing
by anavvanzin
AI-assisted video editing workflows for cutting, structuring, and augmenting real footage. Covers the full pipeline from raw capture through FFmpeg, Remotion, ElevenLabs, fal.ai, and final polish in Descript or CapCut. Use when the user wants to edit video, cut footage, create vlogs, or build video content.
e2e-testing
by anavvanzin
Playwright E2E testing patterns, Page Object Model, configuration, CI/CD integration, artifact management, and flaky test strategies.
dmux-workflows
by anavvanzin
Multi-agent orchestration using dmux (tmux pane manager for AI agents). Patterns for parallel agent workflows across Claude Code, Codex, OpenCode, and other harnesses. Use when running multiple agent sessions in parallel or coordinating multi-agent development workflows.
investor-materials
by anavvanzin
Create and update pitch decks, one-pagers, investor memos, accelerator applications, financial models, and fundraising materials. Use when the user needs investor-facing documents, projections, use-of-funds tables, milestone plans, or materials that must stay internally consistent across multiple fundraising assets.
crosspost
by anavvanzin
Multi-platform content distribution across X, LinkedIn, Threads, and Bluesky. Adapts content per platform using content-engine patterns. Never posts identical content cross-platform. Use when the user wants to distribute content across social platforms.
frontend-slides
by anavvanzin
Create stunning, animation-rich HTML presentations from scratch or by converting PowerPoint files. Use when the user wants to build a presentation, convert a PPT/PPTX to web, or create slides for a talk/pitch. Helps non-designers discover their aesthetic through visual exploration rather than abstract choices.
mle-workflow
by anavvanzin
Production machine-learning engineering workflow for data contracts, reproducible training, model evaluation, deployment, monitoring, and rollback. Use when building, reviewing, or hardening ML systems beyond one-off notebooks.
api-design
by anavvanzin
REST API design patterns including resource naming, status codes, pagination, filtering, error responses, versioning, and rate limiting for production APIs.
backend-patterns
by anavvanzin
Backend architecture patterns, API design, database optimization, and server-side best practices for Node.js, Express, and Next.js API routes.
investor-outreach
by anavvanzin
Draft cold emails, warm intro blurbs, follow-ups, update emails, and investor communications for fundraising. Use when the user wants outreach to angels, VCs, strategic investors, or accelerators and needs concise, personalized, investor-facing messaging.
bun-runtime
by anavvanzin
Bun as runtime, package manager, bundler, and test runner. When to choose Bun vs Node, migration notes, and Vercel support.
archive-fallback
by anavvanzin
Retry-and-fallback routine for Gallica/Europeana/LOC/BnF/Numista image fetches. Try IIIF manifest first, 2 retries with backoff, then Playwright bypass, then metadata-only record with #imagem-pendente. Use when the user says "arquivo falhou", "tentar fallback", "gallica 429", "europeana timeout", or when SCOUT or ARGOS reports repeated archive failures. Invoked as a sub-routine by SCOUT/ARGOS.
ssd-health
by anavvanzin
Verify /media/ana/SSD_DATA SSD mount, symlink integrity for data/raw/{BR,FR,UK,DE,US,BE}, optionally ingest new images from the SSD drop zone into corpus/imagens/, and run the SSD backup script. Use when the user says "ssd", "mount check", "verificar ssd", "ingest drive", "backup iconocracia", "sync ssd", or before any corpus operation that reads or writes to data/raw/.
ecc-iconocracy-guide
by anavvanzin
Mapa de tarefas da tese ICONOCRACIA para as skills, modos do agente, comandos ECC e ferramentas MCP realmente disponíveis neste repositório. Use quando o usuário pedir "qual skill uso para X", "que ferramentas tenho", "guia ecc", "como faço Y na tese", ou estiver em dúvida sobre qual entrada acionar.
corpus-scout
by anavvanzin
Agente de pesquisa iconográfica autônoma para o corpus da tese ICONOCRACY (PPGD/UFSC). Localiza candidatos ao corpus em acervos digitais públicos (Gallica, BnF, Europeana, Library of Congress, BNDigital, British Museum, Numista, Colnect) e entrega cada candidato como nota atômica pronta para o vault Obsidian com frontmatter padronizado, análise iconográfica preliminar, classificação de regime iconocrático, e rastreabilidade GitHub/Drive. Acionar sempre que a tarefa envolver: buscar imagens para o corpus, encontrar candidatos por país/período/suporte/regime, auditar lacunas de cobertura do corpus, gerar notas SCOUT-XXX, executar qualquer das 16 campanhas de busca, ou expandir o corpus sem upload manual de imagens. Nunca acionar para análise de imagens já enviadas pelo usuário — nesse caso usar o IconoCode diretamente.
sync-corpus
by anavvanzin
Run the full corpus sync pipeline in safe sequence. Use when the user says "sync", "sincronizar", "pipeline sync", "atualizar corpus", or after adding new candidates to corpus-data.json.
release-gate
by anavvanzin
Run the full public-release gate in safe sequence before any Hugging Face dataset snapshot, public corpus-data.json export, or site release. Use when the user says "release", "gate", "liberar", "publicar dataset", "snapshot HF", "antes de publicar". Stops at the first failure.
ai-agent-super-skill
by anavvanzin
Comprehensive AI agent building skill merging Perplexity Computer's skill creation, webserver, and automation capabilities with Claude Code's agent orchestration, MCP server building, RAG system construction, subagent coordination, parallel agent dispatching, prompt optimization, and execution planning. Covers designing AI agents, building MCP servers, creating RAG pipelines, orchestrating multi-agent systems, optimizing prompts, and deploying AI-powered workflows. Use when building AI agents, creating MCP servers, designing RAG systems, coordinating subagents, optimizing prompts, or architecting any AI-powered automation.
pandoc-fix
by anavvanzin
Use when Pandoc compile of the ICONOCRACIA thesis fails — undefined citation keys, LaTeX package errors, Markdown→LaTeX escape issues (underscores, ampersands, unicode), YAML frontmatter errors, missing .csl/.bib, or broken cross-references. Diagnoses the error, applies minimal fix, re-runs compile.
scout-dedupe
by anavvanzin
Run dedupe verification on a SCOUT candidate before saving to vault/candidatos/. Combines textual check (corpus-dedup agent) with visual CLIP check (iconocracy_clip.py) when an image path is available. Use when the user says "dedupe", "verificar duplicata", "checar duplicata", or BEFORE any "salvar" of a new SCOUT note.
zotero-cite
by anavvanzin
Resolve a thesis citation via Zotero MCP, insert into vault/tese/references.bib if missing, and return the Pandoc @cite-key ready to paste into a chapter. Use when the user says "citar", "cite", "cita", "insere citação", "zotero", or references an author/title while drafting tese/manuscrito/*.md.
abnt-precommit
by anavvanzin
Use before committing thesis chapters (vault/tese/.md, tese/manuscrito/.md) to catch ABNT NBR 6023:2025 violations — missing page refs on direct quotes, malformed (AUTOR, ano) in-text citations, orphan [@key] without matching BibTeX entry, missing "Acesso em:" on URL references. Runs regex gate first, then delegates LLM-level review to abnt-checker agent.