annotate-project

To annotate a codebase, this Skill reads source files and extracts major blocks and sub-blocks, generating a defun and line-tag map for Emacs simply-annotate. It merges the extracted line-based positions with an existing database and validates the resulting thread data for integrity.

The best way to add LLM-driven documentation tags is applying two-tier tagging: defun tags for major syntactical blocks and line tags for notable sub-blocks. This workflow supports multi-language codebases by extracting meaningful structural positions to aid code understanding and onboarding.

Block-tagging works by reading a project's source code and recording line-based positions for major blocks and smaller sub-blocks. These LLM-driven tags enable fast filtering and exploration, keeping narrative prose separate in a sibling .org file to support efficient codebase review and onboarding.

Yes, you can use simply-annotate with an existing project database. The workflow automatically merges newly extracted line-based positions with your existing database, subsequently validating the combined thread data for integrity to ensure no data is lost during the annotation process.

Yes, this codebase annotation approach supports multi-language projects including Java and Python. It extracts major syntactical blocks and meaningful sub-blocks across different languages to generate a navigable simply-annotate database for code review, onboarding, and documentation.

You should not use LLM-driven tags if your project lacks clearly defined major syntactical blocks or if you need to embed narrative prose directly within source files, as this workflow stores project-local narratives separately in a sibling .org file.