# pulse8-ai-cortex-knowledge-vault

This MCP server gives agents a Git-versioned Markdown vault, typed links, and cross-project context search for persistent knowledge.

- MCP server · `type=mcp` · Claude Code, Codex, Cursor, GitHub Copilot, Windsurf (Devin Desktop), Cline, Zed, Gemini CLI · ★ 15
- Author: synpulse8-opensource
- License: Apache-2.0 · policy=link-only · description only; the file stays with the author
- Source: https://github.com/synpulse8-opensource/pulse8-ai-cortex-knowledge-vault
- Updated: 2026-07-27
- HTML: https://affpapa.org/en/nevibecode/mcp/pulse8-ai-cortex-knowledge-vault

## What it does

The server turns a Markdown directory into shared storage for agents and people. Notes can carry typed node kinds, tags, wikilinks, and graph edges, while Git keeps the vault diff-friendly. Agents can retain knowledge across sessions and model changes.

It combines full-text and hybrid retrieval with a MarkItDown-based file compiler. The MCP interface is available through stdio or streamable HTTP, and a matching REST surface lives under `/api/v1/`. The vault can ingest PDFs, DOCX, PPTX, XLSX, HTML files, and images. Docker is required to run it.

An OpenRouter key is needed for LLM-based links between wiki articles; file conversion works without that key. The design targets persistent, graph-based knowledge rather than short-lived chats or plain document search.

## Requirements
- Needs: Python and uv (commands run via uvx)
- Environment variables: `CORTEX_VAULT_PATH`, `CORTEX_LLM_API_KEY` — keys and tokens: ask the human for the values, never commit them or put them in shared files.

## Install

### Claude Code (native platform)

**Platform CLI**
```bash
claude mcp add pulse8-ai-cortex-knowledge-vault -e CORTEX_VAULT_PATH=<CORTEX_VAULT_PATH> -e CORTEX_LLM_API_KEY=<CORTEX_LLM_API_KEY> -- uvx pulse8-ai-cortex-knowledge-vault
```
Replace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from pulse8-ai-cortex-knowledge-vault <command>

### Codex

**Platform CLI**
```bash
codex mcp add pulse8-ai-cortex-knowledge-vault --env CORTEX_VAULT_PATH=<CORTEX_VAULT_PATH> --env CORTEX_LLM_API_KEY=<CORTEX_LLM_API_KEY> -- uvx pulse8-ai-cortex-knowledge-vault
```
Replace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from pulse8-ai-cortex-knowledge-vault <command>

### Cursor

**Config file**
File `.cursor/mcp.json`:
```json
{
    "mcpServers": {
        "pulse8-ai-cortex-knowledge-vault": {
            "command": "uvx",
            "args": [
                "pulse8-ai-cortex-knowledge-vault"
            ],
            "env": {
                "CORTEX_VAULT_PATH": "<CORTEX_VAULT_PATH>",
                "CORTEX_LLM_API_KEY": "<CORTEX_LLM_API_KEY>"
            }
        }
    }
}
```
Replace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from pulse8-ai-cortex-knowledge-vault <command>

### GitHub Copilot

**Config file**
File `.vscode/mcp.json`:
```json
{
    "servers": {
        "pulse8-ai-cortex-knowledge-vault": {
            "type": "stdio",
            "command": "uvx",
            "args": [
                "pulse8-ai-cortex-knowledge-vault"
            ],
            "env": {
                "CORTEX_VAULT_PATH": "<CORTEX_VAULT_PATH>",
                "CORTEX_LLM_API_KEY": "<CORTEX_LLM_API_KEY>"
            }
        }
    }
}
```
Replace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from pulse8-ai-cortex-knowledge-vault <command>

### Gemini CLI

**Platform CLI**
```bash
gemini mcp add -e CORTEX_VAULT_PATH=<CORTEX_VAULT_PATH> -e CORTEX_LLM_API_KEY=<CORTEX_LLM_API_KEY> pulse8-ai-cortex-knowledge-vault uvx pulse8-ai-cortex-knowledge-vault
```
Replace <…> with your own values. uvx runs the command named after the package; if the README names another one, use uvx --from pulse8-ai-cortex-knowledge-vault <command>

## When to use it
- keep agent memory across sessions
- connect Markdown notes as a graph
- retrieve context across projects
- turn PDFs and DOCX files into knowledge
- version the vault with Git

## Example prompts
- Save the decision from this session to the knowledge vault.
- Find notes related to the current task and build the relevant context.
- Convert `report.pdf` into a vault entry and link it to related notes.

## Similar
- [Markdown Vault MCP](https://affpapa.org/en/nevibecode/mcp/markdown-vault-mcp.md) — MCP server · ★ 35
- [Memory Vault](https://affpapa.org/en/nevibecode/mcp/memory-vault.md) — MCP server · ★ 65
- [grounded-vault](https://affpapa.org/en/nevibecode/skills/grounded-vault.md) — skill · 1400 installs
- [agent-knowledge](https://affpapa.org/en/nevibecode/mcp/agent-knowledge.md) — MCP server · ★ 19
- [Blue Synergy Cortex](https://affpapa.org/en/nevibecode/mcp/blue-synergy-cortex.md) — MCP server · ★ 8
- [Hive Vault](https://affpapa.org/en/nevibecode/mcp/hive-vault.md) — MCP server · ★ 11

---
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