haft
Engineering decisions engine that know when they're stale. Frame, compare, decide — with evidence decay and parity enforcement. For Claude Code, Cursor, Gemini
What is haft?
Engineering decisions engine that know when they're stale. Frame, compare, decide — with evidence decay and parity enforcement. For Claude Code, Cursor, Gemini CLI, Codex and more.
Overview of haft
What it is
Haft is a local governance layer for AI coding agents. It provides durable project memory: what problem is being solved, which options were compared, which decisions the human made, what evidence supports them, and what has gone stale. It is built on the First Principles Framework (FPF) by Anatoly Levenchuk.
Key features
- Durable project memory for AI coding agents, tracking decisions and evidence decay - Parity enforcement across multiple AI coding agent hosts - Supports Claude Code, Cursor, Gemini CLI, Codex, Grok, Hermes, Zed, Antigravity, and others - Local-first architecture with SQLite database for structured artifact graph - MCP gates and transformed skills for host integration - Versioned FPF source brought into the agent's working context
Use cases
- Engineering decisions tracking and comparison - Ensuring decisions are not stale in AI-assisted engineering projects - Maintaining a shared project memory across different AI coding agents
Getting started
bash curl -fsSL https://raw.githubusercontent.com/m0n0x41d/haft/main/install.sh | bash After installation, initialize a project: bash haft init # Interactive host multi-select (TTY only) haft init --claude # Claude MCP + skills + CLAUDE.md section See the repository README for full initialization options.
License and pricing
The license is not stated (NOASSERTION). The project is listed as open source.
Based on the project's public README and repository metadata.