Laying out the streets…STEP 1 OF 3

Vexen · Office #63,734

Memory that stops AI agents from repeating failed fixes

In the Working with AI coding agents neighborhood (AI and agents). In town since 6 Nov 2026.

Nobody has claimed this office yet. Tablif records every new product and gives it an office; when its owner claims it, the lights go on.

Is this yours? Claim it

What Tablif read on Vexen's site

Vexen describes itself as: “GitHub - OnurByte/Vexen: Local-first repository memory for coding agents, built in Rust. · GitHub”

What it is

Calls itself “repository memory”.

  • Calls itself: repository memory
“Local-first repository memory for coding agents, built in Rust.”

Where it puts itself

Rules out “cloud or telemetry”.

  • Says it will not: cloud or telemetry
“Your AI agent forgot what failed. Vexen didn't.”
“No Vexen cloud or telemetry.”

Who it sells to

Sells to developers.

  • Sells to: developers
“Local-first repository memory for coding agents, built in Rust.”

How it runs

A command-line tool for Windows. Runs locally and works offline. Distributed through the npm. Open source. AI is a side feature.

  • Kind: cli
  • Runs on: Windows
  • Runs where: on the device, offline
  • Distributed via: npm
  • Open source: yes
  • AI role: a side feature
“…npx @onurbyte/vexen setup”
“On Windows PowerShell use npx.cmd @onurbyte/vexen setup .”

What it works with

Names Claude Code, Codex and 1 more on its pages.

  • Names: Claude Code · Codex · Cursor
“In the first published pilot, Claude Code with Haiku solved 55% of tasks with relevant memory, against 10% without (11 of 20 against 2 of 20).”
“Start a task in Claude Code. Continue in Codex or Cursor. Saved lessons stay with the project.”

What it offers as proof

Live, no waitlist. Claims “55% of tasks with relevant memory, against 10% without (11 of 20 against 2 of 20)”. Publishes docs.

  • Number claimed: 55% of tasks with relevant memory, against 10% without (11 of 20 against 2 of 20)
“In the first published pilot, Claude Code with Haiku solved 55% of tasks with relevant memory, against 10% without (11 of 20 against 2 of 20).”
“…docs -> https://github.com/OnurByte/Vexen/tree/implementation/docs”

Sources

Tablif read these pages, last on 6 Nov 2026. Everything above is what the product says about itself; Tablif does not verify every claim.

Similar products in Tablif Town

Explore Tablif Town