Index Knowledge

This generates AGENTS.md files across your codebase by scoring directories for complexity and creating documentation where it matters most. It fires off parallel explore agents to hunt for patterns, anti-patterns, and conventions, then combines that with bash analysis and LSP symbol data to decide which folders warrant their own knowledge base file. The root always gets full treatment, subdirectories only if they score high enough on metrics like file count, symbol density, and reference centrality. It's built for keeping AI context current without drowning in generic advice. The dynamic agent spawning is clever: bigger projects automatically get more exploration coverage. Use create-new flag to wipe and regenerate from scratch, otherwise it updates existing files and adds new ones where the codebase has grown complex enough to justify them.

skills.sh installs: 1.3K
GitHub stars: 25.2K
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add tursodatabase/turso --skill index-knowledge --agent claude-code

Installs into .claude/skills of the current project.

Files

index-knowledge

Generate hierarchical AGENTS.md files. Root + complexity-scored subdirectories.

Usage

--create-new   # Read existing → remove all → regenerate from scratch
--max-depth=2  # Limit directory depth (default: 5)

Default: Update mode (modify existing + create new where warranted)

Workflow (High-Level)

Discovery + Analysis (concurrent)
- Launch parallel explore agents (multiple Task calls in one message)
- Main session: bash structure + LSP codemap + read existing AGENTS.md
Score & Decide - Determine AGENTS.md locations from merged findings
Generate - Root first, then subdirs in parallel
Review - Deduplicate, trim, validate