Onemind.md – give your repo a memory without any extra tooling

lazardanlucian1 pts0 comments

GitHub - lazardanlucian/onemind.md: onemind.md turns any git repo into a persistent "mind". Thoughts live as git objects on a hidden ref (refs/mind/main) — written via plumbing, invisible to git status. You get decisions with reasons, full history/diff/blame on your thinking, and it all clones with the repo. No database, no service, no API key — just git. · GitHub

/" data-turbo-transient="true" />

Skip to content

Search or jump to...

Search code, repositories, users, issues, pull requests...

-->

Search

Clear

Search syntax tips

Provide feedback

--><br>We read every piece of feedback, and take your input very seriously.

Include my email address so I can be contacted

Cancel

Submit feedback

Saved searches

Use saved searches to filter your results more quickly

-->

Name

Query

To see all available qualifiers, see our documentation.

Cancel

Create saved search

Sign in

/;ref_cta:Sign up;ref_loc:header logged out"}"<br>Sign up

Appearance settings

Resetting focus

You signed in with another tab or window. Reload to refresh your session.<br>You signed out in another tab or window. Reload to refresh your session.<br>You switched accounts on another tab or window. Reload to refresh your session.

Dismiss alert

{{ message }}

lazardanlucian

onemind.md

Public

Notifications<br>You must be signed in to change notification settings

Fork

Star

main

BranchesTags

Go to file

CodeOpen more actions menu

Folders and files<br>NameNameLast commit message<br>Last commit date<br>Latest commit

History<br>5 Commits<br>5 Commits

AGENTS.md

AGENTS.md

LICENSE

LICENSE

ONEMIND.md

ONEMIND.md

README.md

README.md

example.gif

example.gif

View all files

Repository files navigation

ONEMIND.md

Give your repo a memory.

Most projects forget everything between sessions — decisions, rejected ideas, context. ONEMIND.md fixes that. It's a single file you drop into any git repo, and suddenly your project remembers: why you chose X over Y, what you tried and abandoned, what you learned.

No database. No service. No new dependencies. Just git.

Why you'd want this

"Why did we do it this way?" — answered, forever, in git log.

"What did we try and reject?" — recorded so nobody re-litigates dead ends.

"What did the last agent figure out?" — the next agent picks up where it left off.

Decisions with reasoning , not just code diffs.

It travels with the repo. Clone it, you get the mind. Delete it, the mind goes with it.

Try it now

curl -sO https://raw.githubusercontent.com/lazardanlucian/onemind.md/main/ONEMIND.md

That drops the spec into your repo. Tell your agent to read it.

How it auto-bootstraps

When your agent first reads ONEMIND.md, it checks whether the mind exists. If not, it initializes it — one git command, no files on disk, no config changes. From that point on, every session starts by reading the mind and ends by recording what was learned.

Add the linker line to your AGENTS.md so the agent remembers to do this every session:

This project's learning memory is an in-repo git mind on `refs/mind/main`<br>— read `ONEMIND.md` and follow its protocol.

What it feels like

Once the mind is set up, your AI agents (or you) can:

Ask about past decisions:

"Why did we pick PostgreSQL over SQLite?"<br>"What did we decide about authentication?"<br>"What approaches did we reject for the caching layer?"

The agent searches the mind with git log --grep and git grep, finds the reasoning, and gives you the answer with full context.

Record new learnings in real time:

"The rate limiter breaks under 10k concurrent connections — found during load testing."

This gets committed to the mind immediately, so the next session (or the next agent) knows about it without re-discovering it.

Track what didn't work:

"Tried Redis for session storage. Too much ops overhead for our scale. Status: dead."

Dead ideas stay recorded so nobody wastes time on them again.

Follow threads of thinking:

"What's the current thinking on the deployment pipeline?"

The agent traces a series of related mind commits tagged with the same thread, giving you the full evolution of an idea.

How the mind works (briefly)

Everything lives on a hidden git ref (refs/mind/main) — not a branch, never checked out, invisible to git status. Thoughts are plain git commits. No files are written to your working directory.

Two kinds of thoughts:

Message-only (default) — the commit message is the content. Quick notes, decisions, observations.

Blob thoughts — structured documents (like ADRs) stored as git blobs. Searchable via git grep.

Optional trailers give thoughts machine-readable structure:

mind: chose Tailwind over CSS modules

Faster prototyping, consistent design tokens, no naming debates.

Tags: styling, design<br>Decision: accepted<br>Confidence: high

Read the full spec in ONEMIND.md for all the details — threads, notes, ADR templates, sync, multi-person workflows.

Syncing

git push origin refs/mind/main # push the mind<br>git fetch origin refs/mind/main:refs/mind/main # pull the...

mind onemind repo main agent session

Related Articles