data-ai MCP Server
The context spine that 10x's every AI coding session. Live in 8 IDEs (Claude Code, Cursor, Cline, Continue, Aider, Codex, Windsurf, Zed) via npm + OpenVSX + Anthropic plugin directory. 89% measured token reduction. Local SQLite, zero cloud, Apache 2.0.
Discovered via github-topic:mcp and last synced 3mo ago.
1. Install the package
npx engramx
2. Add to claude_desktop_config.json
{
"mcpServers": {
"engramx": {
"command": "npx",
"args": [
"engramx"
]
}
}
}Config file location: ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) / %APPDATA%\Claude\claude_desktop_config.json (Windows)
12,800
Aider only
proprietary
4,500
Closes this gap
✅
Value
VS Code ext
No
Re-injects god nodes and active landmines right before Claude compresses the conversation. Survives compaction.
✅
Continue.dev
Apache 2.0
CartoGopher
unknown
No
`/init`
✅
—
unknown
`/hook-stats`
9,200
Extracts keywords from the prompt, runs a budget-capped pre-query, injects results before the agent responds.
Baseline (tokens)
3,800
Observer only. Writes to `.engram/hook-log.jsonl` for `hook-stats`.
3,200
Decisions, learnings, project context
What you run
feature
14,500
Injects a compact project brief (god nodes, graph stats, top landmines, git branch). Bundles MemPalace context in parallel.
2,800
Source
Auto-switches project context when you navigate to a different repo mid-session.
6,800
Library API docs for detected imports
Integration
Endpoint
5,500
Project notes, architecture docs
Continue @RepoMap
8,200
External providers cache into SQLite at SessionStart. Per-read resolution is a cache lookup, not a live call. If a provider is unavailable it is skipped silently — you always get at least the structural summary. **Plus: any MCP server becomes a provider via a 10-line plugin file** — see [Plugins extend what engram understands](#plugins-extend-what-engram-understands) above. **The 9 hook handlers:**