# growmos

Keeps one shared memory of your repository for you and your agents: answers questions with edge citations, records facts, and fact-checks claims.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 9
- Author: codician-team
- License: MIT · policy=link-only · description only; the file stays with the author
- Source: https://github.com/codician-team/growmos
- Updated: 2026-08-17
- HTML: https://affpapa.org/en/nevibecode/mcp/growmos

## What it does

growmos builds a knowledge graph out of what you already write: docs, ADRs, READMEs, design notes, session logs. Entities become nodes, short verb-phrase relations become edges, and each edge carries its provenance. Content is hashed, so only what changed re-enters the pipeline.

Writes are one-liners — `growmos remember`, `link`, `journal`. New names get matched against the canonical set; anything unmatched turns into a provisional cluster instead of disappearing. `growmos query` serializes the k-hop subgraph around a question and the answer has to cite edge ids, while `growmos check` verifies claims against edges.

No API key is required for the normal path: the CLI handles the deterministic work and hands extraction, resolution and summarization to whichever agent you run, as a task packet with a prompt, a JSON shape and the exact `growmos apply` command. Python ≥ 3.9 is the only prerequisite. Headless mode (`growmos ingest`) covers cron and CI, and that path does call an API.

## Requirements
- Needs: Python and uv (commands run via uvx)

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add growmos -- uvx growmos
```
uvx runs the command named after the package; if the README names another one, use uvx --from growmos <command>

### Codex

**Platform CLI**
```bash
codex mcp add growmos -- uvx growmos
```
uvx runs the command named after the package; if the README names another one, use uvx --from growmos <command>

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "growmos": {
            "command": "uvx",
            "args": [
                "growmos"
            ]
        }
    }
}
```
uvx runs the command named after the package; if the README names another one, use uvx --from growmos <command>

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "growmos": {
            "type": "stdio",
            "command": "uvx",
            "args": [
                "growmos"
            ]
        }
    }
}
```
uvx runs the command named after the package; if the README names another one, use uvx --from growmos <command>

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add growmos uvx growmos
```
uvx runs the command named after the package; if the README names another one, use uvx --from growmos <command>

## When to use it
- turn READMEs and ADRs into a queryable graph
- answer a multi-hop question about the code with edge citations
- record a fact from a session and link it to existing nodes
- fact-check an agent's claim against its sources
- keep the graph current after a commit without running anything by hand

## Example prompts
- Run growmos init in this repo and show me the first task packet
- Take the extraction packet from growmos next and process README.md
- What does the graph say about the Store and who decided that?

## Similar
- [para-memory-files](https://affpapa.org/en/nevibecode/skills/para-memory-files.md) — skill · 1800 installs
- [mempalace-recall](https://affpapa.org/en/nevibecode/skills/mempalace-recall.md) — skill · 1400 installs
- [memory-continue](https://affpapa.org/en/nevibecode/skills/memory-continue.md) — skill · 1300 installs
- [memory](https://affpapa.org/en/nevibecode/skills/orchestkit-memory.md) — skill · 749 installs
- [agentdb-query](https://affpapa.org/en/nevibecode/skills/agentdb-query.md) — skill · 745 installs
- [supermemory](https://affpapa.org/en/nevibecode/skills/supermemoryai-supermemory.md) — skill · 548 installs

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md