AI coding agents spend time and tokens rediscovering the same codebase context. enola. OSS creates a knowledge graph of the codebase and exposes it through MCP, allowing agents to understand systems in seconds. In our benchmark across three repositories using Claude Haiku, with 102k facts, enola. saved 209k tokens, and completed analysis 14x faster. Built for teams using AI development workflows who want lower costs, faster execution, and consistent results. Open source and self-hostable

Hey Product Hunt 👋 We're seeing a new problem emerge as AI agents become part of software development. Every time an agent starts a task, it often re-discovers the same information about the codebase. The result is wasted tokens, slower execution, and duplicated work. We built enola. OSS to give AI agents persistent knowledge about a repository. Instead of repeatedly analyzing the same files, agents can reuse previously discovered facts and focus on the task at hand. In our early benchmark across three repositories with Claude Haiku, enola. reused 102k facts, saved 209k tokens, and completed analysis 14x faster. We're releasing it as open source because we believe the future developer stack will need infrastructure designed specifically for AI agents. Would love your feedback, ideas, and challenges.

Enola. Open Source Launch

Make AI coding agents faster and use fewer tokens

Make AI coding agents faster and use fewer tokens