Skip to content

Gucky/RecursiveIndex

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

3 Commits
 
 
 
 
 
 

Repository files navigation

Recursive Index

Recursive Index Skill

AI Agents - Skill MIT License LinkedIn

Recursive Index is an agent skill for Codex and other AI coding assistants. It establishes and maintains a scoped index.md navigation system in which every participating folder describes only its direct contents. Detail stays local: a child folder has its own index instead of a repeated recursive file tree.

Project instructions decide where this system applies. That makes it suitable for documentation, resources, operational workspaces, and other non-code areas without forcing index.md files into source packages or generated directories.

Installing Recursive Index

Install the skill with npx:

npx skills add https://github.com/Gucky/RecursiveIndex --skill recursive-index

To install it globally for Codex:

npx skills add https://github.com/Gucky/RecursiveIndex --skill recursive-index --agent codex --global

For a specific agent:

npx skills add https://github.com/Gucky/RecursiveIndex --skill recursive-index --agent claude-code

For all supported agents:

npx skills add https://github.com/Gucky/RecursiveIndex --skill recursive-index --agent '*'

Using Recursive Index

In Codex, trigger the skill directly:

$recursive-index

Or ask naturally:

Use Recursive Index to add direct-content indexes to the documentation and resources folders, but not to source code.

The skill helps agents to:

  • create or update lowercase index.md files in selected folder trees;
  • describe files and directories by purpose without producing recursive manifests;
  • keep entries predictable with directories first and files second, alphabetically within each group;
  • respect allowlist or denylist project scope; and
  • keep indexes accurate after structural changes.

Scope and Safety

Define the participating paths in an applicable AGENTS.md or equivalent project instruction file. The skill does not infer that source, dependency, build, cache, generated, or binary directories should receive indexes. It does not reorganize files or modify source code as part of index work.

Requirements

  • Node.js for npx installation
  • An AI coding assistant that supports agent skills

About

An installable skill for Codex and other AI coding assistants that creates and maintains scoped, local index.md navigation for project folders. Each index describes only direct contents, with allowlist or denylist rules that keep source, build, and generated directories out.

Resources

Stars

Watchers

Forks

Releases

Packages

Contributors