Show HN: Replen – maps your repos to a knowledge graph to match open-source (github.com)

Replen has launched a new tool that enhances AI coding workflows by creating a localized intelligence layer that maps user repositories to a personalized knowledge graph. This innovative system allows Codex, Claude Code, and similar AI coding agents to understand what each project does based on its unique capabilities, rather than relying on trending lists or keyword searches. By continuously monitoring the open-source software ecosystem, Replen scores relevant projects against a user's actual coding capabilities and autonomously suggests libraries, algorithms, or techniques that can benefit their development process, all while ensuring that the user's code remains private. This tool is significant for the AI/ML community as it aims to streamline the development process, improve the relevance of suggestions, and foster better integration of tools without overwhelming developers with noise. Key components of Replen include Brainstem, which matches project capabilities with open-source candidates; Watchtower, which maintains a network of over 1,250 sources for necessary updates; and Atlas, the knowledge graph that illustrates the developer's ecosystem. The result is a more efficient coding experience: meaningful suggestions delivered only when genuinely relevant, thus enhancing productivity and maintaining focus on high-impact changes.