claude-meister

Claude_Meister

Intelligence Runtime for Claude Code

Claude_Meister is a lightweight runtime layer that makes Claude Code smarter, cheaper, and faster — without changing how you use it. It installs in under two minutes and works silently in the background from that point on.

---

Before / After

Scenario	Without Claude_Meister	With Claude_Meister	Difference
Typo fix — tokens loaded into context	~1,400 (full CLAUDE.md always loaded)	~847 (slim baseline only)	-40% context overhead
Typo fix — extra runtime cost	0 (paid full CLAUDE.md cost regardless)	0 (LIGHT mode, runtime skipped entirely)	Same task, cheaper
Moderate refactor — context	1,400 (no routing, same cost every time)	847 + ~800 (targeted context loaded)	Only loads what's needed
Complex feature — memory	Manual, no scoring, often too much retrieved	Scored, ranked, 500-token cap enforced	Controlled, relevant memory
Architectural task — finding tools	Manual directory search, minutes	tool_loader.py ranked matches, seconds	Seconds vs. minutes

What is a "token"? A token is roughly 3/4 of a word. Claude Code has a "context window" (a limit on how much text it can consider at once). Every token loaded costs money and uses up that window. Claude_Meister keeps that number low.

---

Table of Contents

1. Why Claude_Meister?
2. Quick Start
3. Plugin Install and Distribution
4. Detailed Installation Guide
5. How It Works
6. Example Walkthroughs
7. Configuration
8. Commands Reference
9. Updating
10. Uninstalling
11. Troubleshooting
12. FAQ
13. Contributing
14. License & Credits

---

Why Claude_Meister?

The Problem

Claude Code loads its instruction file (~/.claude/CLAUDE.md) at the start of every single conversation — a typo fix and a full architectural redesign both pay the same context cost. There is no built-in way to load only what a task actually needs.

On top of that:

• Memory must be managed manually. There is no built-in way to score, rank, or cap retrieved memories.
• Finding the right tool script in a large project requires manual search.
• Nothing tracks which mode Claude is operating in or whether the context spend was worthwhile.

The Solution

Claude_Meister installs three things:

1. Runtime engine (~/.claude_runtime/) — a set of behavioral documents and controller scripts. The router reads your task, classifies its complexity, and loads only the context that complexity warrants.
2. Memory server (~/.claude_memory/) — an MCP (Model Context Protocol — a standard way for tools to communicate with Claude Code) server that stores, scores, and retrieves memories automatically.
3. Wiki knowledge base (~/.claude_wiki/) — optional offline documentation that Claude can query without an internet connection.

Your ~/.claude/CLAUDE.md gets a small block appended that teaches Claude how to invoke the runtime. That is the only change to any file you already have.

Results

Running python install.py --stats shows your own data:

Claude_Meister Usage Report (last 30 days)
-------------------------------------------
Tasks logged:           47
Mode distribution:      LIGHT 62% | STANDARD 30% | DEEP 8%
Avg memory tokens:      287 / 500 cap
Tasks with memory:      14 (30%)
Tasks skipping runtime: 29 (62%)  <-- saved ~600 tokens each

Estimated savings:      ~17,400 tokens saved by LIGHT mode skipping runtime

---

Quick Start

If you are comfortable with the command line and just want to get started:

# 1. Clone the repo
git clone https://github.com/Mintsolester/claude-meister.git
cd claude-meister

# 2. Run the installer
python install.py --full

# 3. Verify everything installed correctly
python install.py --verify

Then restart Claude Code. That is it.

Not sure what any of those commands mean? Read the Detailed Installation Guide below — it walks through every step with screenshots and expected output.

---

Plugin Install and Distribution

If you want to use Claude_Meister as a plugin through Claude Code marketplaces:

# 1. Add this repository as a marketplace
claude plugin marketplace add Mintsolester/claude-meister

# 2. Install the plugin from this marketplace
claude plugin install claude-meister@claude-meister-marketplace

For local plugin development and testing without marketplace install:

claude --plugin-dir ./plugins/claude-meister

Before publishing plugin changes, run validation locally:

claude plugin validate .
claude plugin validate plugins/claude-meister