srk0102/cogram ? reverse-engineered prompt

Build me a Python app that works like an intent aware memory layer for AI agents.

I want it to remember not just facts, but why a fact matters, what goal it supports, and the thinking pattern behind it. The memory should keep track of people, decisions, relationships, and changes over time, then let an agent search it later using both meaning and relationships. Please include a way to store and query this graph based memory, plus a small server so tools like Claude Desktop or Cursor can connect to it and use the memory across sessions.

It would be great if the project also has a simple dashboard or way to inspect the graph, and some caching so repeated lookups are fast. Please make the default setup easy to run locally with Docker or a single command, include sensible example data, and add clear docs for how to ingest memory and query it. Look up current docs online if you need to.