What is the key difference between Engram and Kage?

Engram belongs to the Knowledge & Memory category with local stdio execution, while Kage belongs to the Knowledge & Memory category with local stdio execution.

How do I install Engram or Kage in Claude Desktop or Cursor?

Copy the mcpServers JSON configuration snippet provided on this page into your claude_desktop_config.json or ~/.cursor/mcp.json file and restart your AI client.

Are Engram and Kage free or paid?

Engram operates under a Free / Open Source model (No auth required). Kage operates under a Free / Open Source model (No auth required).

Which MCP server has more tools and community activity?

Engram lists 6 tools with 1 views and 7 GitHub stars. Kage lists 6 tools with 0 views and 32 GitHub stars.

Can I use both Engram and Kage together in the same MCP host?

Yes! MCP hosts like Claude Desktop, Cursor, Windsurf, Cline, and VS Code support configuring multiple servers under the mcpServers object simultaneously.

How do I install Engram or Kage in Claude Desktop, Cursor, or Windsurf?

Both servers follow the standard Model Context Protocol configuration format. Simply copy the JSON block from the configuration section above and paste it into your AI client's mcpServers configuration object (for instance in claude_desktop_config.json or ~/.cursor/mcp.json ), then completely restart or refresh the application.

Are Engram and Kage free to use, or do they require API keys?

Engram is listed under a Free / Open Source model with No auth required . Kage operates under a Free / Open Source model with No auth required . If environment variables are required (such as API keys), be sure to define them under the env key in your MCP client's configuration file.

Can I run both Engram and Kage at the same time in my AI client?

Yes! MCP clients support multi-server orchestration. You can include both hbarefoot-engram and kage-core-kage as distinct keys inside the single mcpServers object in your configuration file. Your AI assistant will automatically select and call the appropriate tool from either server during chat sessions.

Which MCP server has higher directory engagement and quality ratings?

On AllMCPs, Engram has a Quality Score of 59 /100 ( Good ) with 1 views, 0 installs, and 7 GitHub stars. Kage holds a Quality Score of 55 /100 ( Good ) with 0 views, 0 installs, and 32 GitHub stars.

Side-by-Side Model Context Protocol Comparison

Engram vs Kage

In-depth architectural comparison of the Engram and Kage MCP servers. Compare execution transports, security boundaries, tool capabilities, quality scores, and ready-to-paste client installation snippets for Claude, Cursor, Windsurf, and VS Code.

At a Glance & Executive Verdict

Engram

Knowledge & Memory · Local stdio

Quality: 59/100 (Good) | Auth: No auth required

Kage

Knowledge & Memory · Local stdio

Quality: 55/100 (Good) | Auth: No auth required

Verdict Summary: Choose Engram if you need specialized Knowledge & Memory tools running via a local process. Choose Kage if your workspace requires Knowledge & Memory integration with local subprocess execution. Both servers can be configured concurrently in your client's mcpServers manifest.

Which MCP Server Should You Choose?

Choose Engram when:

You need dedicated capabilities in the Knowledge & Memory domain.
You prefer local stdio subprocess transport architecture.
Your security boundary fits: No auth required (Free / Open Source).
Primary tools included: engram_remember, engram_recall, engram_forget.

Explore Engram Details

Choose Kage when:

You need dedicated capabilities in the Knowledge & Memory domain.
You prefer local stdio subprocess transport architecture.
Your security boundary fits: No auth required (Free / Open Source).
Primary tools included: Memory stored as plain JSON in git following Open Knowledge Format, Deterministic verification rejecting stale or hallucinated citations, Local-only operation with no account or API key required.

Explore Kage Details

Feature & Specification Comparison

Specification	Engram HBarefoot Knowledge & Memory	Kage kage-core Knowledge & Memory
Summary	Local-first persistent memory for AI agents. SQLite + local embeddings (all-MiniLM-L6-v2), hybrid semantic + FTS5 recall, secret detection, and contradiction handling. 6 MCP tools (remember, recall, forget, feedback, context, status) over stdio. Zero cloud, no API keys, fully offline. Works with Claude Desktop/Code, Cursor, and Windsurf. npm install -g @hbarefoot/engram	Verified, git-native memory for coding agents. Memory is plain JSON packets committed in your repo, each checked against the code it cites — hallucinated citations rejected at write, stale or changed memory withheld at recall, plus diff-time stale-catch. Local-only (BM25 + vectors), no account, no API key. npx -y @kage-core/kage-graph-mcp install
Category & Scope	Knowledge & Memory	Knowledge & Memory
Quality signal	59/100 (Good)	55/100 (Good)
Transport Protocol	Local Subprocess (stdio)	Local Subprocess (stdio)
Auth Requirement	No auth required	No auth required
Pricing Model	Free / Open Source	Free / Open Source
Required Env Vars	None required	None required
Compatible Clients	Claude DesktopCursorWindsurfClineVS Code	Claude DesktopCursorWindsurfClineVS Code
Install path signal	`npx · high`	`npx · high`
Engagement & Health	1 views 0 copies 0 upvotes 7 stars	0 views 0 copies 0 upvotes 32 stars
Verified / Official	Community Listing	Community Listing
Open full listing	View Engram Listing	View Kage Listing

Tools & Capabilities Breakdown

Engram Tools (6)

engram_remember

Store a memory with category, entity, confidence, namespace, tags. Auto-runs secret detection.

engram_recall

Hybrid semantic + FTS5 search. Supports `category`, `namespace`, `threshold`, and `time_filter`.

engram_forget

Delete a specific memory by ID.

engram_feedback

Vote a memory helpful/unhelpful. Drives the feedback loop above.

engram_context

Pre-formatted context block (`markdown` / `xml` / `json` / `plain`) with a token budget for system-prompt injection.

engram_status

Health check: memory count, model status, configuration.

Kage Tools (6)

Memory stored as plain JSON in git following Open Knowledge Format

Deterministic verification rejecting stale or hallucinated citations

Local-only operation with no account or API key required

Background proxy for zero-wiring integration with Anthropic agents

Per-directory auto-attach configuration for seamless agent routing

Compatibility with multiple coding agents including Claude Code and Codex

Ready-to-Paste Client Configurations

Paste either (or both) of these JSON server blocks into your client config file (e.g. claude_desktop_config.json or ~/.cursor/mcp.json).

Engram Configuration

mcpServers (Claude Desktop / Cursor)

{
  "mcpServers": {
    "hbarefoot-engram": {
      "command": "npx",
      "args": [
        "-y",
        "package"
      ]
    }
  }
}

Kage Configuration

mcpServers (Claude Desktop / Cursor)

{
  "mcpServers": {
    "kage-core-kage": {
      "command": "npx",
      "args": [
        "-y",
        "@kage-core/kage-graph-mcp"
      ]
    }
  }
}

Frequently Asked Questions

Engram is categorized under Knowledge & Memory and uses a local stdio subprocess. In contrast, Kage belongs to Knowledge & Memory using local stdio subprocess. Select Engram when you need capabilities focused on knowledge & memory and Kage when you require tools for knowledge & memory.

More alternatives to Engram More alternatives to Kage Knowledge & Memory category hub