What is jevmem and what problem does it solve?

Long-term memory for AI agents in which Jev makes the fast typed decisions, such as memory safety, kind, contradictions, relevance and sufficiency, so no generative LLM runs in the memory loop between sessions.

How does jevmem use Jev?

Jev answers typed questions with probabilities: is this note safe, what kind is it, does it contradict an older one, which notes matter for this question, and is that enough yet. System Two handles slower reasoning.

What paper does jevmem follow?

The Jev-Mem paper (arXiv 2609.23986), titled 'System-One-Controlled Agentic Memory for Efficient AI Agents', which splits memory work between a fast System One control plane and slow System Two reasoning.

What forms does jevmem ship in?

As an MCP server, Claude Code hooks, a CLI and a Python library, according to the repo description; the README also shows PyPI and CI badges. Python 3.12 or newer is required.

Is jevmem stable enough for production?

No: the README badge marks it beta, with 2 stars, created 2026-10-03 and pushed 2026-10-06. The digest truncates the System Two description and the install section.

What does jevmem cost to run?

The README claims fractions of a cent per operation because memory work runs through Jev's fast decisions rather than a generative LLM. The digest does not include measured cost tables or latency numbers.

How does jevmem differ from other agent memory add-ons?

It is an implementation of the Jev-Mem paper's System-One-controlled design, and unlike memory add-ons that summarise, tag and re-rank with a big LLM on every read and write, it never calls a generative model for memory upkeep.

Jev directory

CommunityJev Agents & MCP

Jevmem — illescasdaniel

Long-term memory for AI agents whose reads and writes are driven by Jev's typed, probabilistic decisions instead of a generative LLM in the loop, inspired by the Jev-Mem paper and shipped as an MCP server, Claude Code hooks, a CLI and a Python library.

See highlights Open github.com

Stars: 2
Language: Python
License: MIT
Reading time: 2 min
Published by: Community
Author: illescasDaniel
Last verified: Oct 2026

illescasdaniel/jev-mem

User repository on GitHub

View on GitHub

Long-term memory for AI agents driven by Jev (System One) decisions: no generative LLM in the loop. MCP server, Claude Code hooks, CLI, Python library.

Forks: 0
Last pushed: Updated Oct 2026

Repo stats from the GitHub API, cached Oct 2026.

This is one of several Jev listings for the same project. The canonical entry is Jevmem — Avinash-jetwani.

Tagscommunity github memory mcp python

Works forJev for Classification Jev for Routing Jev for Verification

Used inJev for Software & DevTools

Languages: Python
Platforms: MCP CLI
Capabilities: Open sourceRequires API key

Highlights

Long-term memory where fast Jev questions (safety, note kind, contradiction, relevance, sufficiency) control the pipeline.
No generative LLM sits in the memory loop; the README claims fractions-of-a-cent costs and self-cleaning memory.
Follows the Jev-Mem paper (arXiv 2609.23986) design of splitting System One control from slower System Two reasoning.
Ships as an MCP server, Claude Code hooks, a CLI and a Python library; jevmem is published on PyPI.
MIT-licensed beta targeting Python 3.12+, with 2 stars, created 2026-10-03 and pushed 2026-10-06.

Watch out

README and GitHub metadata reviewed 2026-10-17; the digest truncates the System Two description and the install section, so setup, the full question set and benchmark numbers are not evidenced here.

Questions about this entry

Short answers drawn from the linked project or page.

Reactions & coverage

Posts, threads, and videos about this entry from around the web.

X: Jev in front of the HeyGen MCP

X: Playwright CLI + Jev vs Playwright MCP

Hacker News: jevmem keeps Claude Code project memory by asking Jev which turns are worth saving.

avinashjetwani61 pointsJevmem – automatic project memory for Claude Code, built on JevRead the thread on Hacker News

X: A local MCP server that cuts computer use from 23s to 0.8s. Cartwright built Cortex, a local MCP server where…

From the Jev guides

Original write-ups on what this entry is for.

MCP tool routing: Jev in front of the tool call

Oct 16, 2026

Tool menus can outgrow the context that reads them. The catalog's Jev-based routers and gates narrow the list: benchmarked tool search, per-tool scoring, fail-closed approvals, and WebMCP in the browser.

Jev Agents & MCP#mcp#tool-routing#agents

Read article

Approval policies for coding agents: hooks, guards, and gateways

Oct 16, 2026

How the Jev guard family turns coding-agent permissions into auto-approve, deny, or ask: hooks that enforce, guards that fail safe, and gateways that gate generation.

Jev Agents & MCP#coding-agents#guardrails#hooks