what is teamlore fr?

lak7/teamlore — explained in plain English

Analysis updated 2026-05-18

★ 6JavaScriptAudience · developerComplexity · 1/5LicenseSetup · easy

vibe map

mindmap
  root((repo))
    What it does
      Shares AI lessons across team
      Stores memory in git folder
      Warns before repeat mistakes
    Tech stack
      JavaScript
      Node.js
      Claude Code hooks
    Use cases
      Avoid repeating team mistakes
      Share tricky bug fixes
      Onboard new teammates faster
    Audience
      Developers
      Engineering teams
    Review
      Agent proposes lesson
      Human merges via PR
      Nothing unreviewed enters

mindmap root((repo)) What it does Shares AI lessons across team Stores memory in git folder Warns before repeat mistakes Tech stack JavaScript Node.js Claude Code hooks Use cases Avoid repeating team mistakes Share tricky bug fixes Onboard new teammates faster Audience Developers Engineering teams Review Agent proposes lesson Human merges via PR Nothing unreviewed enters

what do people make with this?

VIBE 1

Stop a teammate's Claude Code from repeating a bug another teammate already fixed.

VIBE 2

Build a growing, reviewed record of hard lessons learned while building a project.

VIBE 3

Onboard a new engineer whose AI assistant instantly knows the team's past mistakes.

VIBE 4

Review proposed AI generated lessons through a normal pull request before they spread.

what's the stack?

JavaScriptNode.jsClaude Code

how it stacks up fr

	lak7/teamlore	anil-matcha/open-ai-vibe-coding-course	arthurmoorgan/drift
Stars	6	6	6
Language	JavaScript	JavaScript	JavaScript
Last pushed	—	2026-06-30	—
Maintenance	—	Active	—
Setup difficulty	easy	easy	moderate
Complexity	1/5	1/5	2/5
Audience	developer	pm founder	general

Figures from each repo's GitHub metadata at analysis time.

how do i run it?

Difficulty · easy time til it works · 5min

Requires Node 18+ and Claude Code with skills and hooks support.

Use freely for any purpose, including commercial use, as long as you keep the copyright notice.

in plain english

teamlore gives a team of developers who use Claude Code a shared memory, so a lesson one person's AI assistant learns gets passed on to every teammate's assistant instead of staying trapped in one person's session. The idea is simple: if one developer's Claude Code fixes a tricky bug or avoids a mistake, that lesson is saved into a small file, and everyone else who pulls the repository automatically gets access to it too. There is no server and no account to set up. The shared memory lives entirely as small text files inside a .lore folder that gets committed to your git repository, so onboarding a new teammate is just a normal git pull. You install it by running a single command in the root of a git repository, then committing the result. When a teammate's Claude Code is about to touch a file covered by an existing lesson, it shows a short message reminding them what was already learned and why, before they repeat the mistake. Each lesson is stored as one small file, limited to about 120 words, tagged with the file paths it applies to, who learned it, and when it should be reconsidered as possibly outdated. Crucially, new lessons are proposed by the AI agent but must be reviewed and merged by a human through a normal pull request, the same way code changes are reviewed, which is meant to stop a bad or manipulated instruction from silently becoming something every teammate's assistant obeys. The tool collects minimal anonymous usage statistics, which can be turned off with an environment variable, and it is released under the MIT license. It currently supports Claude Code, with support for other coding assistants planned.

prompts (copy fr)

prompt 1

Run npx teamlore init in this repo and explain what files it creates.

prompt 2

Show me an example teamlore entry for a mistake involving a flaky test.

prompt 3

Explain how teamlore decides which stored lessons apply to the file I'm currently editing.

prompt 4

Walk me through reviewing and merging a new lore entry through a pull request.

Frequently asked questions