# NxtSoft Labs > Documentation for NxtSoft Labs open-source developer tools. Featured project — CGraph: A native C++ engine that turns codebases into knowledge graphs for coding agents. Every documentation page below links to its raw Markdown (.md) form — clean, token-cheap, no HTML chrome. The full corpus is at /llms-full.txt. ## CGraph Source repository: https://github.com/taylor009/CGraph ### Introduction - [Overview](https://open.nxtsoft.io/docs/cgraph/overview.md): What CGraph is, the problem it solves, and how its pieces fit together. - [Getting Started](https://open.nxtsoft.io/docs/cgraph/getting-started.md): The shortest path from a checkout to a graph an agent can query. - [Installation](https://open.nxtsoft.io/docs/cgraph/installation.md): Prerequisites and the full build, from clone to a verified binary. - [Quick Start](https://open.nxtsoft.io/docs/cgraph/quick-start.md): Build a graph, run the daemon, and register CGraph with a coding agent. ### Core Concepts - [Core Concepts](https://open.nxtsoft.io/docs/cgraph/core-concepts.md): The vocabulary and mental model behind CGraph — from extraction to the graph an agent queries. - [Graph Model](https://open.nxtsoft.io/docs/cgraph/graph-model.md): What CGraph's nodes and links represent, and how the graph is shaped. - [Memory](https://open.nxtsoft.io/docs/cgraph/memory.md): How CGraph persists the graph, keeps it warm, and serves context to agents. ### Architecture - [Architecture](https://open.nxtsoft.io/docs/cgraph/architecture.md): How CGraph's pieces fit together — from a source tree to a graph an agent queries. - [Indexing Pipeline](https://open.nxtsoft.io/docs/cgraph/indexing-pipeline.md): How CGraph turns a source tree into a resolved, queryable graph. - [Daemon Architecture](https://open.nxtsoft.io/docs/cgraph/daemon.md): The graphd lifecycle — startup, watching, persistence, and shutdown. - [Retrieval](https://open.nxtsoft.io/docs/cgraph/retrieval.md): How a query travels from a client or agent to an answer from the graph. - [Context Packing](https://open.nxtsoft.io/docs/cgraph/context-packing.md): How CGraph gathers a neighborhood and fits it into an agent's token budget. - [Incremental Updates](https://open.nxtsoft.io/docs/cgraph/incremental-updates.md): How the daemon keeps the graph current without rebuilding it. - [Internals](https://open.nxtsoft.io/docs/cgraph/internals.md): The engine at the center, and what is and isn't public about it. ### Integrations - [MCP Integration](https://open.nxtsoft.io/docs/cgraph/mcp.md): How CGraph speaks the Model Context Protocol, and the tools it exposes. - [AI Agent Integration](https://open.nxtsoft.io/docs/cgraph/ai-agents.md): Wiring CGraph into Claude Code, Codex, and Cursor over MCP. ### Reference - [CLI Reference](https://open.nxtsoft.io/docs/cgraph/cli.md): The cgraph, graphd, and cgraph-client commands and their flags. - [Configuration](https://open.nxtsoft.io/docs/cgraph/configuration.md): The environment variables and resolution rules that configure CGraph. ### Guides - [Examples](https://open.nxtsoft.io/docs/cgraph/examples.md): End-to-end workflows — from a fresh checkout to an agent reasoning about your code. ### Performance - [Benchmarks](https://open.nxtsoft.io/docs/cgraph/benchmarks.md): What CGraph measures, how, and the one result published so far. - [Performance](https://open.nxtsoft.io/docs/cgraph/performance.md): How CGraph stays fast, and the levers you have — grounded in mechanics, not invented numbers. ### Project - [Design Decisions](https://open.nxtsoft.io/docs/cgraph/design-decisions.md): The choices that shape CGraph, and the reasoning and tradeoffs behind each. - [Contributor Guide](https://open.nxtsoft.io/docs/cgraph/contributing.md): How to build, test, and contribute to CGraph. - [Roadmap](https://open.nxtsoft.io/docs/cgraph/roadmap.md): Where CGraph is heading — direction, not dated promises. - [FAQ](https://open.nxtsoft.io/docs/cgraph/faq.md): Common questions about what CGraph is, and isn't. - [Changelog](https://open.nxtsoft.io/docs/cgraph/changelog.md): Released versions of CGraph — as they are cut.