Mengram

Memory that survives /clear, compaction and the next agent

103 followers

Memory that survives /clear, compaction and the next agent

103 followers

Visit website

LLM Memory

Open-source memory for Claude Code, Codex, Cursor and your own agents. Hooks recall on every prompt, save after each turn, checkpoint before compaction, and hand the task to the next session or agent with a card: done, remaining, last test and the commit it ran on. Every fact says which tool wrote it and when. 23-62x less context than sending the history, with the numbers and the losses published. Apache-2, works without an account for the checkpoint and the task card.

This is the 2nd launch from Mengram. View more

Mengram 2.47

Launched this week

Memory that survives /clear, compaction and the next agent

Seven months after the first launch, this one is about the work, not memory types. Hooks for Claude Code, Codex and Cursor: recall on every prompt, save after each turn, a checkpoint before compaction that puts the working state back. New: `mengram resume`, a task card (done, remaining, last test and its commit) the next session or agent picks up, keyed by repo and branch. Every fact says which tool wrote it and when. Benchmark: 23-62x fewer context tokens than sending history, losses published.

Free Options

Launch tags:Open Source•Developer Tools•Artificial Intelligence

Launch Team

Customer.ioAutomate Messaging Everywhere — Startups Get 12 Months Free

Promoted

Mengram

Hunter

📌

Hi PH, Ali here, solo maker. I launched Mengram seven months ago as "an AI memory API with three types". That was the wrong pitch: nobody buys memory types. What people actually lose is the work. Claude Code compacts and forgets which files it just edited, a new session re-explains the project, Cursor and Codex don't know what Claude decided yesterday, and the next agent redoes a step that was already done. So this version is about that, and it comes with numbers instead of adjectives. What's new since March: - Hooks for Claude Code, Codex and Cursor: recall on every prompt, save after each turn, a checkpoint before compaction that puts the working state back (files you edited, last commands, your last request). - `mengram resume`: a task card written when an agent stops (done, remaining, last test and the commit it ran on), keyed by repo + branch so another session, another agent or another machine picks it up. - Every fact says where it came from (tool, session, date), and a Mac path never reaches a Linux session. - A salience gate before write: 45% less junk stored, no recall lost in 18 runs. - A benchmark you can re-run: over 90 simulated days, 23-62x fewer context tokens than sending the whole history, same or better recall on personal facts. Where it still loses, published: on customer-support dialogue at 30 days recall was 0.25 until this week; the bug was my contradiction pass archiving a loyalty number for a vaguer restatement. It's 0.875 now, and the experiment log with the rejected results is in the repo (experiments/QUEUE.md, RESULTS.jsonl). Install: pip install mengram-ai && mengram setup --key . Or no account at all: the checkpoint and the task card work locally. Apache-2. Question for you: what do you lose most between sessions, the state, the decisions, or the reasons? That's what I'll build next.

Report

4d ago

Previous Mengram Launches

MengramAI memory API with 3 types: facts, events, and workflows

Launched on February 19th, 2026