by m0n0x41d
Structured reasoning framework for Claude Code, Gemini, Cursor, and Codex — hypothesis-driven decision making with auditable evidence trails
# Add to your Claude Code skills
git clone https://github.com/m0n0x41d/quint-codeStructured reasoning for AI coding tools — make better decisions, remember why you made them.
Supports: Claude Code, Cursor, Gemini CLI, Codex CLI
Works exceptionally well with Claude Code!
Instead of jumping to conclusions and solutions, your AI generates competing hypotheses, checks them logically, tests against evidence, then you decide. Everything is in your plain sight.
No more archaeology in chat history. Decisions live in .quint/ —
queryable, auditable, yours.
curl -fsSL https://raw.githubusercontent.com/m0n0x41d/quint-code/main/install.sh | bash
Or build from source:
git clone https://github.com/m0n0x41d/quint-code.git
cd quint-code/src/mcp
go build -o quint-code .
sudo mv quint-code /usr/local/bin/
cd /path/to/your/project
quint-code init
This creates:
.quint/ — knowledge base, evidence, decisions.mcp.json — MCP server configuration~/.claude/commands/ — slash commands (global by default)Flags:
| Flag | MCP Config | Commands |
|------|-----------|----------|
| --claude (default) | .mcp.json | ~/.claude/commands/*.md |
| | | |
| | | |
| | * | |
| | All of the above | All of the above |
| | — | Commands in project dir instead of global |
--cursor.cursor/mcp.json~/.cursor/commands/*.md--gemini~/.gemini/settings.json~/.gemini/commands/*.toml--codex~/.codex/config.toml~/.codex/prompts/*.md--all--local* Codex CLI limitation: Codex doesn't support per-project MCP configuration. Run
quint-code init --codexin each project before starting work to switch the active project in global codex mcp config.
/q0-init # Initialize knowledge base
/q1-hypothesize "Your problem..." # Generate hypotheses
Here is a library of some workflow examples that might help you kick off with probing.
But really, it would be better to hack into it straight away and feel the flow. Shash commands have a numeric prefix for your convenience.
For best results, we highly recommend using the CLAUDE.md from this repository as a reference for your own project's agent instructions. It's optimized for software engineering work with FPF.
At minimum, copy the FPF Glossary section to your:
CLAUDE.md (Claude Code).cursorrules or AGENTS.md (Cursor)This helps the AI understand FPF concepts like L0/L1/L2 layers, WLNK, R_eff, and the Transformer Mandate without re-explanation each session.
Quint Code implements the ...