Profile My skills Saved Collections Edit profile Submit a skill

← all publishers

amgrbi96

@amgrbi96 source repo

10 published skills

  1. ▌
    Openwa · amgrbi96 bundle
    OpenWA self-hosted WhatsApp API Gateway — full operational and reference guide. Covers install (Docker/OrbStack), boot, pairing a WhatsApp number via QR, enabling the MCP server (full mode, 39 tools), minting API keys, wiring Claude/Cursor/ZCode clients, calling tools, on/off, the REST API, webhooks + Socket.IO events, configuration, the 3-role auth model, SDKs, architecture, and 13 gotchas. Use when the user says "set up openwa", "boot openwa", "install openwa", "pair my whatsapp number", "enable openwa mcp", "add openwa to zcode/claude/cursor", "send a whatsapp message via openwa", or wants to drive WhatsApp programmatically. Also use when working in the OpenWA repo, code hits port 2785, builds the openwa-api Docker service, sends X-API-Key headers with owa_k1_ keys, subscribes to message.received/message.ack/session.status webhooks, or toggles DATABASE_TYPE/QUEUE_ENABLED/REDIS_ENABLED/ENGINE_TYPE.
    0
    installs
  2. ▌
    PDF Tools · amgrbi96 bundle
    PDF output-side engineering for what pymupdf cannot do: HTML-to-PDF generation with Puppeteer (print CSS, web fonts, React templates), PDF/A archival compliance via ghostscript and verapdf, PKCS#7 digital signing with @signpdf, and qpdf encryption/repair/linearization. NOT for content extraction — use the parse-docs skill. Routine page edits (merge, split, rotate, crop, watermark, forms, redaction) need no skill — use pymupdf directly.
    0
    installs
  3. ▌
    Flashcards · amgrbi96 bundle
    Create high-yield study flashcards from any source material (PDF book, lecture notes, markdown, plain text). Use when: (1) the user wants flashcards / Anki cards / a study deck from a document or topic, (2) turning a chapter/lecture/article into review cards, (3) capturing a single concept or fact as a flashcard to add to a deck, (4) exporting cards to Anki (.apkg) or Markdown. Emits Markdown for review then optionally Anki. Enforces one-fact-per-card atomicity and a high-yield filter so output is condensed, not exhaustive.
    0
    installs
  4. ▌
    Prisma CLI · amgrbi96 bundle
    Use when running or configuring the PRISMA Literature Review Pipeline, constructing CLI commands, setting up literature review runs, choosing screening providers, configuring discovery sources, interpreting output artifacts, or creating per-project config files. Also use when the user mentions systematic reviews, literature screening, paper discovery, PRISMA flow, citation snowballing, or anything related to academic paper search and analysis — even if they don't explicitly name the tool.
    0
    installs
  5. ▌
    Mineru · amgrbi96 bundle
    Parse PDFs, Word docs, PPTs, and images into clean Markdown using MinerU's cloud VLM engine. Use when: (1) Converting PDF/Word/PPT/image to Markdown, (2) Extracting text/tables/formulas from documents with highest accuracy, (3) Batch processing multiple files, (4) Saving parsed content to Obsidian or knowledge bases. Supports LaTeX formulas, tables, images, multilingual OCR, multi-token quota pooling, resume, and dry-run validation. Cloud-based — needs internet + API token.
    0
    installs
  6. ▌
    Prospero Search · amgrbi96 bundle
    Search the PROSPERO registry for registered systematic reviews and check for duplicate or overlapping reviews before registering a new one. Use when the user mentions PROSPERO, systematic review registration, checking for duplicate reviews, or asks whether their research question has already been studied — even if they don't explicitly say 'PROSPERO'. Also triggers on mentions of 'protocol registration', 'review overlap', 'CRD number lookup', or 'prospective register'. Useful for researchers planning systematic reviews, scoping reviews, or meta-analyses who need to verify novelty before registration.
    0
    installs
  7. ▌
    Liteparse · amgrbi96
    Parse PDFs, Office documents (DOCX/PPTX/XLSX/ODT), and images locally into text, Markdown, or JSON with bounding boxes and OCR confidence — no cloud, no LLM, nothing leaves the machine. Use when the user wants local/offline/private document parsing or OCR (scanned PDFs, photos, screenshots), spatial text coordinates, page screenshots, or batch conversion of mixed-format folders — even casually ("grab the text from this", "read this image", "what does this scan say"). If the user needs highest accuracy (formulas, complex layouts, degraded scans) and cloud is acceptable, route to mineru instead.
    0
    installs
  8. ▌
    Parse Docs · amgrbi96 bundle
    Routes document parsing to the right tool — pdf-to-markdown for fast Markdown, pymupdf-pdf for local layout/tables, liteparse for multi-format OCR/tables, or mineru for cloud VLM accuracy. Three modes: single document (decision tree shortlist, then the user picks), folder batch (per-file routing + resume), or all-methods (every applicable tool per document). Use this skill before parsing any document (PDF, DOCX, PPTX, XLSX, images) — it checks which tools are installed, shortlists the fitting ones, and asks the user to choose, with a quick comparison in each option. Use when the user mentions parsing, extracting text, converting a document, OCR, tables, drug dosing data, or batch processing — even casually ("grab the text", "read this", "pull content from").
    0
    installs
  9. ▌
    Pymupdf PDF · amgrbi96 bundle
    Local PDF workbench on PyMuPDF — extract text/tables/images to Markdown or JSON, merge/split/rotate/delete pages, render to PNG, read/write metadata and TOC, search, encrypt/decrypt, plus recipes for annotations, forms, redaction, and PDF creation. Use for any local, no-cloud PDF task; prefer mineru for VLM-grade accuracy and pdf-tools for JS/Puppeteer pipelines.
    0
    installs
  10. ▌
    PDF To Markdown · amgrbi96 bundle
    Extract text from PDFs as structured, semantic Markdown. Use when converting a PDF to Markdown, extracting text from a PDF, processing one or more PDFs into Markdown output, reading PDF contents for analysis, ingesting documents for RAG pipelines, preparing PDFs for LLM context, or any task where PDF text needs to be in a machine-readable format. ALWAYS use this skill when the user has a PDF and needs its content as text or Markdown — even if they don't explicitly say "convert to markdown". Fastest option for PDFs with a text layer; cannot OCR scanned PDFs on the free tier (route those to mineru/liteparse).
    0
    installs