Local LLM
roam-code
Local codebase intelligence CLI and MCP for coding agents.
★ 517 starsApache-2.0
View guide →// local-llm · bartolli/codanna
codanna
codanna is an open-source local llm project by bartolli (bartolli/codanna on GitHub). It has 744 GitHub stars and 70 forks, is released under the Apache-2.0 license, and was last updated on GitHub on 24 September 2026.
Local code-intelligence MCP server and CLI for coding agents.
💎 Hidden gem★ 744 stars⑂ 70 forksApache-2.0updated 2026-09
▦ visual — this project's README ships no embeddable screenshot,
so here's a placeholder. The guide below is complete regardless.
so here's a placeholder. The guide below is complete regardless.
// project facts
Project facts
- Repository
- bartolli/codanna
- GitHub stars
- 744
- Forks
- 70
- License
- Apache-2.0
- Last updated on GitHub
- 24 September 2026
- Category
- Local LLM
- Listing
- Hidden gem — a shorter page for a lesser-known project
- Official site
- Official docs
// how it works
How it works
- Install codanna
Get the binary per the docs.
- Index your repo
Build the local code index.
- Connect your agent
Expose it as an MCP server.
- Query the code
Agents search call graphs and symbols.
// features
Why people love it
Code indexing
Builds a local index of your repo.
Call-graph analysis
Agents trace callers and callees.
MCP server
Plugs into any MCP-capable agent.
CLI included
Query the index from the shell.
Local-first
No code leaves your machine.
Apache-2.0
Free for commercial use.
Advertisement
// run it
Run it yourself
- Step 1 — Clone the repo
Download the source and enter the folder.
git clone https://github.com/bartolli/codanna.git cd codanna - Step 2 — Build it
Compile the Rust binary.
cargo build --release - Step 3 — Index a repo
Point it at your codebase.
./target/release/codanna index /path/to/repo
// use cases
What you can build
- Give coding agents a searchable code index.
- Trace call graphs without manual grepping.
- Keep code intelligence fully local.
- Speed up agent onboarding to big repos.
Advertisement
// faq
Questions, answered
What is codanna?
A local code-intelligence MCP server and CLI for AI coding agents, at bartolli/codanna.
How do agents use it?
As an MCP server exposing code search and call graphs.
Does it upload my code?
No. Indexing is local-first.
Is it free?
Yes. Apache-2.0 licensed.
Where are the docs?
At docs.codanna.sh, linked from the repo.
// keep exploring
Similar tools
Dev Tools
CodeGraphContext
An MCP server plus a CLI tool that indexes local code into a graph database to provide context to AI assistants.
★ 4.2k starsMIT
View guide →Local LLM
code-context-engine
Code-context-engine, in Python — built for save 94% on ai coding tokens. index your cod...
★ 425 starsMIT
View guide →Local LLM
agentdock
Secure MCP runtime letting AI agents operate your local machines.
★ 1.2k starsApache-2.0
View guide →Local LLM
ownmem
Open source, Git-native memory for AI coding agents - deterministic local recall
★ 423 starsApache-2.0
View guide →Local LLM
marm-memory
A local-first 3-in-1 AI memory layer and MCP server.
★ 416 starsApache-2.0
View guide →// alternatives
Alternatives in Local LLM
The most-starred other Local LLM projects in the LibreAI catalogue — the realistic shortlist if codanna is not the right fit.
- Ollama★ 182.1k
The simplest way to run open language models on your own machine.
- Open Webui★ 153.9k
User-friendly AI Interface (Supports Ollama, OpenAI API, ...)
- llama.cpp★ 130.2k
LLM inference in C/C++
- vLLM★ 93.1k
A high-throughput and memory-efficient inference and serving engine for LLMs
- lobehub★ 83.0k
🤯 LobeHub is your Chief Agent Operator, organizing your agents into 7×24 operations by hiring, scheduling, and reporting on your entire AI team.
// more in local-llm
More in Local LLM
Nearby Local LLM projects in the catalogue, closest in stars to codanna — keep exploring the shelf either side of it.
- pasteguard★ 761
AI gets the context.
- rustunnel★ 659
Rustunnel, in Rust — built for self-hosted, secure tunnel server in rust. expose local.
- comfyui-mcp★ 785
Agent-native control plane for ComfyUI — generate media in plain chat.
- gpt-home★ 651
ChatGPT for your home, running on a Raspberry Pi.
- can-i-finetune-this★ 792
Check if your GPU can fine-tune an LLM — memory estimates and LoRA recipes.
- Agent★ 642
Agent! now supports macOS 14.6 or later, Apple Silicon and Intel, latest release.
Get codanna
Free and open — grab the source, star the repo, and start creating.