Back to directory

pi-memory

A persistent memory extension for the Pi coding agent, keeping durable facts, a daily log, and a scratchpad as plain Markdown files, with optional semantic search across everything remembered.

pi-memory

pi-memory gives the Pi coding agent memory that survives between sessions, since an agent otherwise forgets everything once a session ends. It keeps durable facts and decisions, a running daily log, and a scratchpad of things to return to, all as plain Markdown files under a memory directory that a user can read, edit, and commit like any other project file.

Six core tools, for writing, forgetting, restoring, and reading memory, using the scratchpad, and checking memory status, work immediately after installation with no further setup. An optional integration with a separate semantic-search tool adds keyword, semantic, and hybrid search across everything the agent has ever remembered; when present, it automatically creates the needed search collection and keeps vector embeddings current in the background, including after the very first install, when the embedding model itself has to download before semantic search is ready.

More in Agent memory

See all 450
  • A long technical guide to AI agent memory: why stateless LLMs need it, the different memory architectures from vector stores to knowledge graphs, and how Cognee implements one as an open-source memory layer.

  • A local, file-based long-term memory layer for a personal AI that never summarizes conversations, keeping every line with a timestamp and searching by time first and words second, rather than retrieving only similar-seeming chunks.

  • NEOTH

    deepwiki.com

    A local-first personal AI daemon in Rust with five memory tiers, consent-gated tools, a WASM plugin sandbox, multi-provider model routing and signed audit logs, aimed at people who want both a daily assistant and a serious coding helper.