
Agent Magnet
- 13 repo stars
- Updated July 22, 2026
- helinakdogan/magnet-gateway
Agent Magnet is a behavioral memory layer MCP that learns preferences and context across AI tools so agents can retain user-specific behavior between sessions.
About
Agent Magnet is an MCP server that acts as a behavioral memory layer for AI tools. It learns your preferences and context across Claude, Cursor, Codex, and similar agents so later sessions start closer to how you already work. A developer reaches for it when repeated re-explaining style, stack choices, or workflow habits slows agent-assisted coding. It is agent infrastructure rather than app feature code, deployment control, or analytics.
- Behavioral memory layer for AI coding tools
- Learns preferences and context across sessions
- Works across Claude, Cursor, and Codex
- Reduces repeated preference explanations to agents
Agent Magnet by the numbers
- Data as of Jul 24, 2026 (Skillselion catalog sync)
claude mcp add --env MAGNET_USER_ID=YOUR_MAGNET_USER_ID --env MAGNET_OPENAI_KEY=YOUR_MAGNET_OPENAI_KEY --env MAGNET_LOCAL_MODE=YOUR_MAGNET_LOCAL_MODE --env MAGNET_REDIS_URL=YOUR_MAGNET_REDIS_URL --env MAGNET_PROJECT_ID=YOUR_MAGNET_PROJECT_ID agent-magnet -- uvx agent-magnetAdd your badge
Show developers this MCP server is listed on Skillselion. Paste this into your README.
| repo stars | ★ 13 |
|---|---|
| Package | agent-magnet |
| Transport | STDIO |
| Auth | Required |
| Last updated | July 22, 2026 |
| Repository | helinakdogan/magnet-gateway ↗ |
How can my AI coding agents remember my preferences and context across Claude, Cursor, and Codex without me repeating them every session?
Persist behavioral memory and preferences across Claude, Cursor, and Codex so agents remember your context.
Who is it for?
Developers using multiple AI coding agents who want persistent preference and context memory across tools.
Skip if: Teams that only need one-off prompts, static project docs, or deployment and infrastructure management instead of agent memory.
What you get
Agents access stored behavioral memory and preferences, producing more consistent responses aligned with your prior interactions.
Recommended MCP Servers
FAQ
What problem does Agent Magnet solve?
It provides a behavioral memory layer so AI tools can learn and reuse your preferences and context across sessions.
Which AI tools does it support?
It is designed to work across Claude, Cursor, and Codex.
What kind of data does it store?
It stores behavioral preferences and context learned from your interactions with supported AI tools.