Gives AI coding agents persistent architectural memory of codebases.
Product memo
Hokmah provides AI coding agents with a persistent architectural memory of codebases, building a live reasoning graph directly from Git history. This approach helps agents understand codebase structure and potential impact, allowing them to find complex bugs that traditional methods often miss. It integrates with existing developer workflows and supports multiple LLMs, making it a specialized tool for AI-assisted development.
For who
AI coding agents and developers
Solves what
Provides AI coding agents with persistent architectural memory of codebases.
- Architectural memory via git history
- Live reasoning graph
- Risk analysis before push
In their own words
AI Code Agent · Architectural Memory · Git-history Native
360,000 stars. Thousands of contributors. Hokmah found what none of them could.
AI Agent with Architectural Memory. One MCP server for every AI coding agent. Analyze impact, generate tests, write code — all from the graph. Free to start.
Commercial cues
Model
subscription
Free tier
Yes
Trial
No
Pricing Strategy
- • A free tier offers basic analysis to drive initial adoption.
- • Custom Enterprise pricing handles unlimited usage for large organizations.
Operator context
Founded
May 2026
Platform
Web app
Audience
Developers
Builder Strategy
- Strategy Type
- Niche Specialist
- Stage
- Pre Revenue
- Effort
- Solo Buildable
About Hokmah Expand
Hokmah equips AI coding agents with a persistent architectural memory, allowing them to understand and interact with codebases more effectively. It builds a live reasoning graph from Git history, enabling features like risk analysis before code pushes and the generation of tests.
Designed for developers and AI coding agents, this tool helps identify complex bugs and improve code quality. The "Bring Your Own Key" (BYOK) model addresses privacy concerns, making it a practical option for teams integrating AI into their development workflows.