Self-learning pattern memory for AI coding agents. Observe patterns, track confidence, promote to rules, and suggest across sessions — with SQLite persistence and MCP integration.
AI agents forget learned patterns between sessions. Every conversation starts from scratch — no memory of what worked, what failed, or what you prefer.
AI agents forget learned patterns between sessions. Repeated mistakes, lost preferences, no continuity.
Observe → track → promote pipeline with SQLite persistence and MCP integration. Patterns gain confidence over time.
Open-source on PyPI, reusable across Claude Code, Cursor, and Windsurf. Agents learn and remember.
Patterns flow through a confidence-based promotion pipeline. Observed patterns gain confidence with each occurrence and automatically mature into actionable rules.
Four pattern types cover the full spectrum of agent behavior — from tool sequences to user preferences.
Track sequences of tools/actions that commonly follow each other. e.g. seq:read->edit->test
Remember user style choices and conventions. e.g. pref:commit_style=conventional
Capture solutions to problems that keep coming back. e.g. fix:import-order-stdlib-first
Things that are commonly used as a pair. e.g. combo:pytest+coverage
Everything an AI agent needs to build persistent memory — no configuration, no external services, no dependencies.
All patterns stored in ~/.instinct/instinct.db. Survives restarts, works offline, zero setup.
First-class MCP server. Works out of the box with any MCP-compatible AI agent or editor.
Works with Claude Code, Cursor, and Windsurf. All agents share the same pattern database.
Use from the command line or import as a Python library. Both interfaces, same database.
Fingerprint-based scoping ensures patterns are relevant to the current project context.
Pure Python, stdlib only. No external packages required. Install and run anywhere.
Simple CLI for observing, suggesting, consolidating, and listing patterns.
Multiple AI agents can share the same instinct database. Claude Code as primary, Cursor or Windsurf as secondary — all reading and writing to the same SQLite file.
instinct-mcp is live on PyPI. The path to universal agent memory.
Published on PyPI. MCP native server, CLI, Python library. SQLite persistence with project-scoped fingerprinting.
Insights dashboard showing pattern trends, promotion rates, and agent learning curves over time.
Share instinct databases across teams. Export/import rules, sync patterns between developers.
More agent integrations, more pattern types, community-contributed rule packs.
Install from PyPI. Check the source. Start learning.