Back to Apps

OpenLore
by clay-good
Persistent architectural memory for AI agents, turning codebases into queryable knowledge graphs to eliminate context decay.
0 stars
Works in:Cursor
Exposes:ToolsResources
What it does
OpenLore provides AI coding agents with structural cognition by mapping a codebase into a knowledge graph. It eliminates the need for agents to repeatedly re-read source files to understand architecture, drastically reducing orientation token costs and preventing "amnesia" between sessions.
Tools
orient(): The primary entry point; returns relevant functions, call paths, and spec domains for a given task.detect_changes: Risk-scores changed functions based on call graph centrality.search_code: Performs semantic and keyword search across the codebase graph.get_subgraph: Retrieves a local neighborhood of the call graph for specific entities.analyze_impact: Determines the blast radius of a proposed change across the system.
Installation
{
"mcpServers": {
"openlore": {
"command": "openlore",
"args": ["mcp"]
}
}
}
Supported hosts
Claude Code, Cursor, Cline, Continue
Quick install
npm install -g openlore && openlore installInformation
- Pricing
- free
- Published
- 6/2/2026
- stars
- 0
Categories
Choose your AI client and follow the steps below.
Cursor
Add to ~/.cursor/mcp.json: {"mcpServers": {"openlore": {"command": "openlore", "args": ["mcp"]}}}Claude Desktop
{"mcpServers": {"openlore": {"command": "openlore", "args": ["mcp"]}}}





