CodeGraph
colbymchenry/codegraph
Local code knowledge graph that gives coding agents, Hermes Agent included, indexed context in one call
CodeGraph is a local code knowledge graph that connects to coding agents, Hermes Agent among them, through an MCP server. It indexes a project's symbols, call edges and dependencies so an agent can fetch the code it needs in one call instead of many grep and read steps.
What CodeGraph does
CodeGraph builds a pre-indexed knowledge graph of every symbol, call edge and dependency in a project and stores it in a local .codegraph/ directory. The README describes the problem it targets: agents usually discover structure by running grep, glob and file reads one at a time, which costs many tool calls and round trips before real work starts. CodeGraph answers those questions from the graph instead, and it keeps the graph current by watching the project and updating on every file change.
Setup has three parts. First the CLI is installed, then codegraph install detects supported agents and wires the CodeGraph MCP server into each one. Hermes Agent is on the list of agents it configures automatically, next to Claude Code, Cursor, Codex CLI, opencode, Gemini CLI, Antigravity IDE, Kiro and GitHub Copilot. Finally codegraph init builds the graph for a given project. The README states that everything runs locally.
Key features
- Pre-built graph of symbols, call edges and dependencies for a project
- Auto-sync that watches the project and updates the graph on every file change
- codegraph install detects supported agents, Hermes Agent included, and adds the CodeGraph MCP server to each
- Standalone installer that needs no Node.js, with an npm package as an alternative
- codegraph upgrade, codegraph uninstall and codegraph uninit commands for updating and removal
- Runs entirely on the local machine
When to use it
- Giving Hermes Agent a code map of a large repository so it spends fewer tool calls exploring
- Sharing one code index across several coding agents installed on the same machine
- Keeping an agent's view of a repository current while it edits files
Who it is for: Developers who run Hermes Agent or other coding agents on sizeable codebases and want them to query an index rather than search files one by one.
How it fits with Hermes Agent
Hermes Agent is one of the agents that codegraph install detects and configures. The tool is general and supports many agents, so it is a compatible tool rather than a Hermes-only one.
How to install CodeGraph
These commands are copied from the project's README. Check the repository for the latest steps before you run them.
curl -fsSL https://raw.githubusercontent.com/colbymchenry/codegraph/main/install.sh | sh
codegraph install
cd your-project
codegraph initRequirements: No Node.js for the standalone installer (npm is an alternative); codegraph install once to connect agents, then codegraph init in each project
FAQ
What is CodeGraph?
CodeGraph is a local code knowledge graph for AI coding agents. It indexes the symbols, call edges and dependencies of a project and exposes them to agents through an MCP server.
Does CodeGraph work with Hermes Agent?
Yes. The README lists Hermes Agent among the agents that codegraph install detects and configures, wiring the CodeGraph MCP server into it.
Is CodeGraph free and open source?
The repository is released under the MIT license. The README also mentions a hosted CodeGraph platform in early beta, which is a separate product from this repository.
Similar dev tools for Hermes Agent
All dev toolsCLI and dashboard that tracks token usage and cost across AI coding agents, including Hermes Agent
liustack ModLensVision plugin that lets text-only coding agents read pasted images as structured JSON evidence
liaohch3 claude-tapLocal proxy and trace viewer that shows the real API traffic of AI coding agents, Hermes Agent included
alexgreensh Token OptimizerAudits wasted tokens in AI coding agents, saves work across compaction, and includes a Hermes installer
coulsontl AI ToolboxDesktop app that manages configs, MCP servers and skills across AI coding tools, including Hermes Agent
greyhaven-ai autocontextSelf-improving harness that runs agent tasks against evaluation and keeps playbooks, traces and datasets
Related guides: How to install Hermes Agent · Run multiple Hermes agents with profiles