https://github.com/maxkle1nz/m1nd

Structural retrieval for coding agents: faster orientation, narrower scope, more precise changes.
https://github.com/maxkle1nz/m1nd

ai-agents ai-tools code-analysis code-graph code-intelligence connectome developer-tools graph-database hebbian-learning knowledge-graph mcp mcp-server neuro-symbolic plasticity rust spreading-activation static-analysis

Last synced: 3 months ago
JSON representation

Structural retrieval for coding agents: faster orientation, narrower scope, more precise changes.

• Host: GitHub
• URL: https://github.com/maxkle1nz/m1nd
• Owner: maxkle1nz
• License: mit
• Created: 2026-03-12T15:11:14.000Z (5 months ago)
• Default Branch: main
• Last Pushed: 2026-04-03T05:18:01.000Z (4 months ago)
• Last Synced: 2026-04-03T08:01:45.945Z (4 months ago)
• Topics: ai-agents, ai-tools, code-analysis, code-graph, code-intelligence, connectome, developer-tools, graph-database, hebbian-learning, knowledge-graph, mcp, mcp-server, neuro-symbolic, plasticity, rust, spreading-activation, static-analysis
• Language: Rust
• Homepage: http://m1nd.world
• Size: 48.5 MB
• Stars: 8
• Watchers: 0
• Forks: 1
• Open Issues: 8
• Metadata Files:
  • Readme: README.de.md
  • Changelog: CHANGELOG.md
  • Contributing: CONTRIBUTING.md
  • Funding: .github/FUNDING.yml
  • License: LICENSE
  • Security: SECURITY.md

Awesome Lists containing this project

• awesome-cli-coding-agents - m1nd - first MCP server that gives a CLI coding agent a neuro-symbolic code graph over the repo: `impact`/`why`/change-reasoning across a call + cross-file-imports graph (10 languages), calibrated-trust retrieval that returns `abstain`/`insufficient_evidence` instead of guessing, and persistent agent memory. A SessionStart hook pre-orients the agent before its first turn. Works with Claude Code, Codex, Cursor, and any MCP client. Rust, MIT. (Harnesses & orchestration / Agent infrastructure)