{"slug": "onemind-md-give-your-repo-a-memory-without-any-extra-tooling", "title": "Onemind.md – give your repo a memory without any extra tooling", "summary": "Onemind.md introduces a single-file protocol that stores project decisions, rejected ideas, and context as hidden git commits on a dedicated ref, enabling AI agents and developers to retrieve and record reasoning without external databases or services.", "body_md": "**Give your repo a memory.**\n\nMost 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.\n\nNo database. No service. No new dependencies. Just git.\n\n**\"Why did we do it this way?\"**— answered, forever, in`git log`\n\n.**\"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.\n\n```\ncurl -sO https://raw.githubusercontent.com/lazardanlucian/onemind.md/main/ONEMIND.md\n```\n\nThat drops the spec into your repo. Tell your agent to read it.\n\nWhen your agent first reads `ONEMIND.md`\n\n, 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.\n\nAdd the linker line to your `AGENTS.md`\n\nso the agent remembers to do this every session:\n\n```\nThis project's learning memory is an in-repo git mind on `refs/mind/main`\n— read `ONEMIND.md` and follow its protocol.\n```\n\nOnce the mind is set up, your AI agents (or you) can:\n\n**Ask about past decisions:**\n\n\"Why did we pick PostgreSQL over SQLite?\" \"What did we decide about authentication?\" \"What approaches did we reject for the caching layer?\"\n\nThe agent searches the mind with `git log --grep`\n\nand `git grep`\n\n, finds the reasoning, and gives you the answer with full context.\n\n**Record new learnings in real time:**\n\n\"The rate limiter breaks under 10k concurrent connections — found during load testing.\"\n\nThis gets committed to the mind immediately, so the next session (or the next agent) knows about it without re-discovering it.\n\n**Track what didn't work:**\n\n\"Tried Redis for session storage. Too much ops overhead for our scale. Status: dead.\"\n\nDead ideas stay recorded so nobody wastes time on them again.\n\n**Follow threads of thinking:**\n\n\"What's the current thinking on the deployment pipeline?\"\n\nThe agent traces a series of related mind commits tagged with the same thread, giving you the full evolution of an idea.\n\nEverything lives on a hidden git ref (`refs/mind/main`\n\n) — not a branch, never checked out, invisible to `git status`\n\n. Thoughts are plain git commits. No files are written to your working directory.\n\nTwo kinds of thoughts:\n\n**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`\n\n.\n\nOptional trailers give thoughts machine-readable structure:\n\n```\nmind: chose Tailwind over CSS modules\n\nFaster prototyping, consistent design tokens, no naming debates.\n\nTags: styling, design\nDecision: accepted\nConfidence: high\n```\n\nRead the full spec in [ ONEMIND.md](/lazardanlucian/onemind.md/blob/main/ONEMIND.md) for all the details — threads, notes, ADR templates, sync, multi-person workflows.\n\n```\ngit push origin refs/mind/main                # push the mind\ngit fetch origin refs/mind/main:refs/mind/main  # pull the mind\n```\n\nThe mind lives on the same remote as your code. CI pipelines should ignore `refs/mind/*`\n\n, so pushing the mind should not trigger a build, etc.\n\n- ONEMIND.md instructs agents not to save PII/TOKENS, sensitive data in the mind.\n- Nothing leaks into your working tree — no\n`mind/`\n\nfolder, no`.gitignore`\n\nedits. - Never commit secrets, tokens, or PII to the mind.\n- Append only. Use\n`git notes`\n\nto correct past thoughts without rewriting history.", "url": "https://wpnews.pro/news/onemind-md-give-your-repo-a-memory-without-any-extra-tooling", "canonical_source": "https://github.com/lazardanlucian/onemind.md", "published_at": "2026-07-12 09:48:32+00:00", "updated_at": "2026-07-12 10:05:21.937646+00:00", "lang": "en", "topics": ["developer-tools", "ai-tools"], "entities": ["Onemind.md", "lazardanlucian"], "alternates": {"html": "https://wpnews.pro/news/onemind-md-give-your-repo-a-memory-without-any-extra-tooling", "markdown": "https://wpnews.pro/news/onemind-md-give-your-repo-a-memory-without-any-extra-tooling.md", "text": "https://wpnews.pro/news/onemind-md-give-your-repo-a-memory-without-any-extra-tooling.txt", "jsonld": "https://wpnews.pro/news/onemind-md-give-your-repo-a-memory-without-any-extra-tooling.jsonld"}}