MervinPraison

Ask natural-language questions about a loaded model

Compare two BioModels entries

Fetch BioModels metadata with get_modelinfo

Scan a parameter across a range

Look up linked publications

Rank search results by relevance

Summarise SBML structure and annotations

Validate SBML XML structure

Edit PraisonAIPPT YAML decks in examples, keep style consistent, regenerate matching PPTX, and verify output. Use when users request slide/verse/section/highlight updates for deck YAML files.

Step-by-step daily_single video build with modular QA gates, spoken↔visual sync, social comparison clips, pytest, and validate-sync. Use for Fable-style pilots, social-comparison videos, A/V sync fixes, or end-to-end daily-single orchestration.

Builds YouTube-style daily single-topic videos from create-news handoff — sync-assets HD crawl, hook montage (≥5 heroes), plain-language beats, ffmpeg assembly, script-locked captions, pixel visual audit, validate-sync ×3. Use for Fable-style stories,…

Builds per-segment HeyGen + ElevenLabs roundup videos from research handoff, PraisonAIPPT avatar_media_3 decks (50590 pattern), align-cues/yaml/build/normalize-audio/merge, validate-all (audio_loudness), HD 1080p compositor output, and mer.vin publish. Use…

Builds on-point video SRT from locked segment scripts plus Whisper timing only — never raw transcription text. Use for daily_single captions, final.srt, segment.srt, subtitles, transcript alignment, YouTube hook/overview/Let's get started structure, or when…

Upload an MP4 to mer.vin WordPress and embed it in a published article with a Gutenberg wp:video block. Use when the user provides a local video path, asks to add video to an article, praisonaiwp media upload, wp media import, or wp:video on mer.vin.

Generate and edit images with OpenAI gpt-image-2 for PraisonAI PPT slide assets. Use when creating diagrams, charts, or visuals for YAML decks (slide_type image), or any "generate/create/edit image" request for this repo.

Downloads YouTube audio and crops one or more timestamp ranges to m4a files only — no transcription, normalisation, or articles. Use when the user gives a YouTube URL with start/end times, asks to crop clips, extract audio segments, or says they only want…

Full local sermon audio pipeline — crop, remove ranges, silence cut, volume normalise, transcribe (Tamil+English auto-detect), extract-text, sanity checks. Run ALL steps when user provides audio; do not stop at transcribe alone. YouTube download →…

Downloads YouTube sermon audio, crops by timestamps, optimises volume (quiet or hot), merges multi-part clips, transcribes with praisonai-audio-editor, maps to YAML decks, and packs deliverables. Use when the user gives a YouTube URL with start/end times,…

Process and extract information from PDF documents. Use this skill when the user asks to read, analyze, or extract data from PDF files.