Codebase Memory vs Plur

Two MCP servers from our memory ranking, compared on live data - updated nightly, never sponsored.

Bottom line · 2026-08-25

Codebase Memory records significantly more monthly package installs (39,800 vs 8,736); both are actively maintained, with commits inside the last month.

How they differ in kind

Codebase Memory focuses on: The fastest and most efficient code intelligence engine for AI coding agents. Plur focuses on: Persistent, open memory for AI agents - local-first, zero-cost, shared across MCP tools (Claude Code, Hermes, OpenClaw, Cursor). Implementation languages differ (C vs TypeScript).

What each one does

Codebase Memory

The fastest and most efficient code intelligence engine for AI coding agents. Full-indexes an average repository in milliseconds, the Linux kernel (28M LOC, 75K files) in 3 minutes. Answers structural queries in under 1ms. Ships as a native executable with a small verified runtime-asset set for macOS, Linux, and Windows - download, run install, done.

From the project's README.

Plur

Persistent, open memory for AI agents - local-first, zero-cost, shared across MCP tools (Claude Code, Hermes, OpenClaw, Cursor). Your agent's memory is plain-text engrams you can read, correct, and delete - not weights you can't. Its 15 documented tools cover plur, learn.

From the project's README.

14 signals, side by side

Signal Codebase Memory Plur
Monthly installs 39,800 8,736
GitHub stars 40,198 241
Last commit 2026-08-23 2026-08-23
Releases · last 90 days 8 10+
In the registry since Feb 2026 Jul 2026
Registry versions 13 10
Documented tools not extracted - see README 15
Env vars extracted none extracted - see README none extracted - see README
Runs local (stdio) local (stdio)
Endpoint auth local only local only
Maintenance actively maintained actively maintained
License MIT Apache-2.0
Language C TypeScript
Category rank #1 of 259 #3 of 259

What their tools cover

Extracted from each project's README - the tools each server hands your agent. A missing list means the README doesn't document one, not that no tools exist.

Codebase Memory no tool list in README

Plur 15 tools documented

  • plur_learn - Store a correction, preference, or convention
  • plur_learn_batch - Store many engrams in one call (batch dedup + per-item failure…
  • plur_recall - Retrieve relevant memories - hybrid (BM25 + embeddings) by default…
  • plur_inject_hybrid - Select engrams for current task within token budget
  • plur_feedback - Rate relevance (trains quality over time)
  • plur_forget - Retire a memory (activation decays, eventually pruned)
  • plur_rescope - Move an existing engram to another scope - personal → team, or back
  • plur_session_scope - Change the session's default write scope mid-session

+7 more on the full page

Which one, for what

Derived from the signals above, not hands-on testing.

Pick Codebase Memory if…

  • → you want the more widely installed option - 39,800 monthly installs vs 8,736
  • → you prefer a C codebase to extend or audit

Pick Plur if…

  • → you prefer a TypeScript codebase to extend or audit
  • → its documented toolset fits - 15 tools including plur_learn, plur_learn_batch

Who's behind them

Codebase Memory - deusdata: 1 MCP server tracked, 1 maintained, 40,198 combined stars.

Plur - plur-ai: 1 MCP server tracked, 1 maintained, 241 combined stars.

Not sold on either?

The next-ranked memory servers we track:

Quick answers

Is Codebase Memory better than Plur?

Package installs favor Codebase Memory: 39,800 monthly installs to 8,736. Plur still ranks #3 of 259 maintained memory servers. Data as of 2026-08-25; we haven't hands-on tested either.

Can I use Codebase Memory and Plur together?

Yes - MCP clients accept multiple servers in one config, so you can enable both memory servers side by side. If their tools overlap, keep the one whose toolset fits to keep your agent's tool list lean.

Keep exploring: best memory servers · TypeScript servers · all comparisons · every server

Methodology: package-usage signal = npm/PyPI installs (last month, platform APIs) · maintenance = commit recency + release cadence (GitHub) · tool lists and environment-variable names extracted from each project's README · endpoint auth from our own nightly probes. We haven't hand-tested these servers; everything here is data as of 2026-08-25.