Convert video talks with slides into annotated blog posts with synchronized transcripts.
npx skills add https://github.com/rajshah4/my-agent-skills --skill annotate-talk
Please help me install this Agent Skill. Skill: annotate-talk Source: https://github.com/rajshah4/my-agent-skills/tree/main/skills/annotate-talk Command: npx skills add https://github.com/rajshah4/my-agent-skills --skill annotate-talk
This skill makes it easy to convert technical talks and slide decks into readable, annotated blog posts with embedded slide images and synchronized transcripts.
Use the ai-annotate-talk CLI to generate a blog post from a YouTube talk, with slides.pdf and output_images/ as inputs: ai-annotate-talk "https://youtu.be/VIDEO_ID" slides.pdf output_images/ ai-annotate-talk "https://youtu.be/VIDEO_ID" slides.pdf output_images/ --output post.md ai-annotate-talk "https://youtu.be/VIDEO_ID" slides.pdf output_images/ --transcript custom.txt ai-annotate-talk "https://youtu.be/VIDEO_ID" slides.pdf output_images/ --prompt context.txt
High-intent search queries and answers about installing and using this skill.
To convert a video talk into an annotated blog post, you provide a YouTube URL and a slides.pdf file to generate a markdown file with synchronized transcripts and embedded slide images. The tool aligns the video audio timeline with the slide deck to produce written content.
To generate synchronized transcripts from a video presentation, you need a YouTube video URL, a corresponding slides.pdf file, and an output directory for extracted slide images. You also need to set the Gemini_API_KEY and JINA_READER_KEY environment variables.
Yes, you can use a custom transcript file when turning a conference talk into a blog post by passing the --transcript flag with your text file. This overrides the default transcript extraction and ensures your written content matches your specific text annotations.
To add specific context when converting a tutorial video to written content, use the --prompt flag with a context.txt file. This injects custom instructions into the annotation process, tailoring the generated blog post to your specific documentation or educational requirements.
Automating video-to-blog conversion requires the Gemini_API_KEY and JINA_READER_KEY environment variables, along with poppler-utils and hamel_tools installed on your system. These dependencies handle PDF processing, web reading, and AI-driven annotation to output the final markdown post.
szeyuGenerate slide deck images from markdown and merge them into PPTX and PDF.
mahg-esGenerate static websites from Markdown into deployable HTML with CI workflows.
ombhanushaliiiTranscribe audio files into text using Whisper Large V3 models.
costiashTranscribe local videos and YouTube content into timestamped transcripts using gpt-4o-transcribe.
