KiroGraph: Local code knowledge graph for AI, optimized for token efficiency
KiroGraph is a semantic code knowledge graph designed for Kiro, enhancing token efficiency by reducing tool calls and enabling instant symbol lookups. It utilizes a local SQLite database to store an indexed representation of the codebase, ensuring that all operations remain on the user's machine. The system supports various indexing layers and can integrate with other MCP-capable tools, although full compatibility is not guaranteed.
- ▪KiroGraph is built specifically for Kiro's MCP and hooks system, offering full support for Kiro only.
- ▪The knowledge graph allows Kiro to query relationships and hierarchies in the codebase without scanning files, resulting in faster responses.
- ▪Indexing is performed using tree-sitter to parse source files, and the data is stored locally in a SQLite database.
Opening excerpt (first ~120 words) tap to expand
KiroGraph Semantic code knowledge graph for Kiro: fewer tool calls, instant symbol lookups, 100% local. Inspired by CodeGraph by colbymchenry for Claude Code, rebuilt natively for Kiro's MCP and hooks system. Full support is for Kiro only. Experimental integrations for other MCP-capable tools (Claude Code, Codex) are available but not fully tested. See Other Tools (Experimental) for details. Why KiroGraph? When you ask Kiro to work on a complex task, it explores your codebase using file reads, grep, and glob searches. Every one of those is a tool call, and tool calls consume context and slow things down. KiroGraph gives Kiro a semantic knowledge graph that's pre-indexed and always up to date.
…
Excerpt limited to ~120 words for fair-use compliance. The full article is at GitHub.