ngvoicu

Kluris — AI skill for Claude Code

AI community

Turn AI agents into team SMEs.

How to install Kluris

This entry records only its repository, not the path inside it, so there is no exact command to give. Open ngvoicu/kluris and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Kluris does

Turn AI agents into team SMEs. Kluris builds a shared, human-curated knowledge brain (decisions, constraints, incidents, tribal knowledge) as markdown in a git repo. Queryable via slash commands from Claude Code, Codex, Cursor, Windsurf, Cline, Gemini CLI.

Alternatives in AI

WeKnora — Open-source LLM knowledge platform: turn raw documents into a queryable RAG, an autonomous reasoning agent, an 26.9k ★
ARIS Agent Guide — For AI agents reading this repo 6.2k ★
Omnara — by Ishaan Sehgal - A command center for AI agents that syncs Claude Code sessions across terminal, web, and mo 2.6k ★

README

Kluris

Kluris turns your AI agents into a subject matter expert that never sleeps and never quits.

*When your best engineer sleeps, Kluris doesn't. When they leave, Kluris stays.*

🎥 **New here?** Take the [guided tour at kluris.ngvoicu.dev](https://kluris.ngvoicu.dev/presentation.html) — install, first brain, agent workflows, multi-brain, git collaboration, and the MRI visualization, end to end.

What is Kluris?

Kluris gives every AI agent on your team shared knowledge -- architecture, decisions, conventions, learnings -- so they work like an SME who knows your entire codebase, not a generic assistant starting from scratch every time.

Knowledge is stored in a **brain**: a git-backed repo of structured markdown that agents read, search, and apply automatically. The human and agent curate the brain together -- the agent proposes what to document, the human reviews and approves every piece.

Why it saves tokens

A brain is a pre-digested summary your agent reads instead of the raw files. Without it, the agent crawls the whole repo (and every sibling repo it needs context from) on every new chat. With a brain, it loads one compact snapshot and jumps straight to the neuron it needs.

**Without kluris** — illustrative cold-start on a medium repo:

tree + README + CLAUDE.md         ~3,000 tokens
grep for related symbols          ~2,000 tokens
read 4-8 relevant files           ~15,000 tokens
read 2-3 sibling-project files    ~8,000 tokens
                                  ─────────────
                                  ~28,000 tokens just to orient

**With kluris** — same task, brain-backed:

kluris wake-up --json snapshot    ~1,200 tokens (brain.md + lobes + recent + glossary)
kluris search "" --json    ~400 tokens   (ranked hits with snippets)
read 1-2 matching neurons         ~1,500 tokens
                                  ─────────────
                                  ~3,100 tokens — ~9x less context burned on orientation
``