4 Kinds of Agent Memory

Post by Paweł Huryn on LinkedIn — “My agent has 4 kinds of memory. 3 live in files.”

The 4 Memory Types

Memory TypeWhat It IsWhere It Lives
Working MemoryWhat the agent holds right now — context window, goals, scratchpadYou manage it (context window)
Semantic MemoryFacts & knowledge — your strategy, rules, what “good” looks likeMarkdown files, versioned in git
Episodic MemoryWhat happened before — past runs, decisions, outcomesA table (CSV in git, not a database)
Procedural MemoryHow to do things — reusable steps loaded on a triggerSkills — also files

“Three of those are just files in git: two folders of markdown, one table. The fourth is the context window. You probably don’t need a vector database either.”

Two More (Escape Hatches)

When there’s too much to read:

  • Vector store — searches a million documents by similarity. Enterprise search, not for your agent.
  • Graph database — relationships with labels (owns, depends on) at scale. But your linked markdown is already a graph the agent walks by reading.

Key Takeaway

“That’s my setup: markdown files plus one CSV. No vector DB, no embeddings, no server. It sharpens every week because the agent writes its own lessons back into the files.”

Most agent memory doesn’t need a vector DB — files in git are enough.