quint-code Documentation

quint-code is a decision engineering tool for AI coding assistants. It helps you frame problems, compare options fairly, record decisions as contracts, and know when those decisions go stale.

It works as an MCP server — a backend that your AI tool (Claude Code, Cursor, Gemini CLI, Codex) connects to. You interact with it through slash commands in your AI tool's chat interface. The agent does the heavy lifting; you make the calls.

What it does

When you face a decision that's hard to reverse, affects the team, or has real consequences if you get it wrong — migrating a database, picking an auth architecture, designing a public API, restructuring a service boundary — you type /q-reason and describe the problem. The agent:

1. Frames the problem (what's broken, what are the constraints, how do we know it's solved)
2. Defines comparison criteria before generating options (prevents bias)
3. Generates genuinely different variants (not three versions of the same idea)
4. Compares them on a Pareto front (no single-score collapse)
5. Records the decision as a contract with invariants, rollback plan, and expiry date

Every decision gets an R_eff score — a computed trust metric that degrades over time as evidence expires. When R_eff drops below 0.5, the decision surfaces as stale. You review it, re-validate, or supersede it with a new decision. This is the the continuous loop of decide → measure → reframe.

Who is this for

Anyone who makes decisions with real consequences — and needs those decisions to be traceable, challengeable, and aware of their own expiry date. Software architects, engineering managers, systems engineers, tech leads. The problems aren't limited to code: technology selection, infrastructure strategy, process design, organizational trade-offs, vendor evaluation, migration planning.

The current interface is developer-focused (MCP server for AI coding tools), but the underlying methodology is domain-agnostic. quint-code aims to implement as much as possible of FPF (First Principles Framework) by Anatoly Levenchuk — a formal approach to structured reasoning that works wherever decisions need evidence, comparison, and lifecycle management. No prior knowledge of FPF is required; quint-code implements the practical parts and stays out of your way.

Supported tools

What's new in 5.3

• Interactive terminal dashboard — quint-code board gives you real-time project health without starting an agent session. Four tabs, vim navigation, glamour markdown, live refresh.
• Context-aware reasoning — /q-reason now reads your intent: think-and-respond (no artifacts), prepare-and-wait (you drive the cycle), or full autonomous (agent drives). No longer defaults to running the full cycle.
• Codebase awareness — 6 languages (Go, JS/TS, Python, Rust, C/C++), module detection, dependency graph, decision coverage, drift detection
• Cross-project recall — decisions from other projects surface when framing new problems
• Decision integrity — adversarial verification, inductive measurement gates, evidence supersession, search keyword enrichment
• Decisions as test specs — full-cycle decisions contain enough structure for any agent to generate property-based tests
• Roadmap — team collaboration, formal specification, and more

Next step

Install quint-code — takes about 30 seconds. Already on 5.0? Migration guide.