Skip to main content

Roadmap

Done (v0.1.1 — current)​

  • Core generation pipeline (TypeScript, ESM, Commander)
  • Mock, OpenAI, Anthropic, Gemini providers
  • 11 output modes: docusaurus, blog, docs, gitbook, readme, faq, troubleshooting, lesson, social, changelog, seo
  • Video Intake Guard — URL validation, tech classification, transcript requirement
  • Transcript Intake Agent — clipboard, stdin paste, file import (.md/.txt/.vtt/.srt), stubs for OAuth + local media
  • inspect-url and validate-source CLI commands
  • transcript command with 5 subcommands
  • --url and --force flags on generate
  • npm run demo one-command demo in mock mode
  • npm run batch for directory-level processing
  • npm run verify output validation
  • npm run doctor setup diagnostics
  • Profile customization (modes/_profile.md)
  • Zod config schema with video_intake and transcript_intake sections
  • 82-test Vitest suite (Node 22/24, Ubuntu + macOS)

In progress / good first issues​

  • npm run preview — local browser preview of generated docs (planned — previous PR closed, open for contribution)
  • Ollama provider — local LLM support (#1)
  • OpenRouter provider — access to 200+ models (#2)
  • Mermaid diagram mode — architecture diagrams from transcripts (#3)
  • Better Docusaurus frontmatter — smarter tag and sidebar_label extraction (#4)

Planned (v0.2.0)​

  • Vimeo metadata provider
  • --watch mode — regenerate on file change
  • Multi-file merging — combine several transcripts into one doc
  • Output diffing — show what changed between two generation runs
  • Prompt library — community-contributed mode prompts

Ideas / RFC​

These are not committed but worth exploring:

  • GitHub Action — run generate in CI when a transcript file is pushed
  • VS Code extension — generate docs from the editor command palette
  • Docusaurus plugin — auto-build from a transcripts/ folder at build time
  • Interactive mode — guided prompt to select mode, provider, and output format

Not planned​

These are explicitly out of scope:

  • YouTube transcript scraping or downloading
  • Automatic publishing to Docusaurus / GitBook / Notion
  • Hosted service or SaaS version
  • GUI / desktop app

Want to contribute to any of these? See Contributing or open an issue.