# lodos vs Obsidian

Source: https://lodos.md/en/compare/obsidian
Language: en

The same local-first principles. A much bigger job.

---

Obsidian is the best notebook there is, and lodos is not trying to beat it at that. lodos is what happens when the notes stop being the whole job: the same markdown and the same graph, plus an encrypted vault, a kanban board, scheduled workflows and an agent that reads all of it and can spend a credential without ever seeing it. Point it at the vault you already have and it reads every page, every link and every folder, without touching a single file.

## The three apps that never quite talk to each other

Obsidian holds the note. Your password manager holds the key the note is about. Something else holds the task the note is for. Each one is good at its job, and not one of them can see the other two.

So you become the integration. You read the runbook, open the vault, copy the key, do the work, come back, write down what happened and move the card. Every step goes through you, because you are the only thing that has all three in context at once.

lodos is those three in one place, with an agent that has all of it in context and none of your keys.

## What Obsidian does better than lodos

This is the comparison where the other tool is genuinely excellent, and pretending otherwise would waste your time.

- A plugin ecosystem nothing in this space comes close to. If a workflow exists in your head, someone has probably already built it.
- A plain-file vault that opens in any editor on any machine, with no application in between.
- Years of refinement as a writing tool. If your job is to think in text, that is exactly what it was built for.

## lodos is not a better Obsidian. It is the half that comes after the notes.

Everything Obsidian is good at, lodos keeps: local files, markdown, links, a graph. What it adds is the machinery that turns a written page into work that actually happens, and a team layer that shares all of it without ever handing a server a key.

### A vault, not a folder of plain files

An Obsidian vault is markdown on disk, which is exactly why it is portable and exactly why an API key does not belong in it. In lodos a secret is encrypted on your device before it touches the disk, with a key per section and a hash-linked record behind every access. The page references it as {{secrets.stripe.sk_live}} and the value lives somewhere the page never does.

### A wiki that maintains itself

Ask a question and get a written answer with [[wikilink]] citations rather than a list of search hits. Imported sources are write-once and hash-checked at the database level, so any claim traces back to where it came from. A nightly pass runs five checks in plain SQL, no AI cost: orphans, dead links, claims older than ninety days, unanswered TODO markers and pages nobody filed. That is the part a folder of files cannot do for you.

### A graph the agent walks, not just one you look at

Same idea, different consumer. One resolver serves the renderer, the linter and the AI, so the agent traverses slugs, titles and edges to build context across several hops, and never page bodies. You still get both views, a local one around the page you are reading and a global one you can filter by type, confidence or orphan. The difference is that something other than you can read the map.

### A board that finishes itself

Kanban tasks the agent reads, plans and moves. When a job starts repeating it becomes a declarative workflow: drafted by the AI, approved by you, running on a schedule, calling your own APIs with credentials injected at the moment of use. Pin the result to your dashboard as a number, a chart or an overdue flag.

### A team layer that never hands the server a key

This is the part with the most engineering behind it. Every member holds their own sealed copy of a section key, and the coordinating server relays addressed envelopes it cannot open, with no delete endpoint to censor with. Invites are confirmed out of band with a safety number, so a coordinator sitting in the middle is detectable rather than trusted. A member without a grant never receives the ciphertext at all: the boundary is cryptographic, not a filter in the interface. Remove someone and the key rotates, the content is re-encrypted and everyone remaining moves to a new epoch, so the envelope they kept opens nothing written after they left. Every action lands in a signed, hash-linked record with a Merkle root you can anchor to an external timestamp, and it records whether a human or an agent did it. Shared, audited, and still zero-knowledge.

### Playbooks instead of plugins

Twenty-four skills ship in the box: weekly metrics, refunds, ADRs, post-mortems, spec authoring, security review. Write your own and a scanner checks it before it can be saved, every time, with no setting to turn that off. A skill is instructions and never code, the app decides which of its seventy-three tools each one may call, and forty-one checks fail the build if anyone tries to add an execution path. It is a narrower deal than an ecosystem, and a much smaller thing to trust.

### One app where you had three

Drop in a PDF, a contract or a spreadsheet and it becomes searchable knowledge. Query your own numbers and get a chart. Ask for a board update and get a rendered deck, offline. The AI chat remembers what you told it last week, on your machine, and the transcript is never written to disk.

### Every machine you own, and your phone

Nine dollars a month for the whole install, not per seat, buys the one thing a cloud is genuinely good at: an encrypted snapshot that brings a new machine up in minutes and keeps the rest in step. Our servers hold ciphertext and an integrity chain, never a key, so a breach on our side is a pile of noise. The desktop runs on macOS, Windows and Linux, and the phone companion reads that same snapshot behind Face ID, so you can check a credential or approve a run without being at your desk.

## Bringing your Obsidian vault across

1. Point lodos at the folder your vault already lives in. No export step: they are the same markdown files.
2. Pages, folder structure and [[links]] come across intact, and a scan refuses anything carrying a plaintext secret so it can go in the vault instead.
3. Ask the agent what is in there. It builds the index, answers with citations, and tells you which pages have already gone stale.

lodos reads that folder, it does not move or rewrite it. Your vault sits exactly where it sat this morning and still opens in Obsidian tonight, which means trying this costs you nothing but the download. Pages stay markdown end to end, so they leave the same way they arrived.

## What can it see?

Both products are described with the same four answers. There is no kinder phrase available for the same behaviour, so neither column can be graded on a curve.

| What can it see? | Obsidian | lodos |
| --- | --- | --- |
| Where your data lives | On your device | On your device, sync optional |
| What the vendor can read | There is no server | Only ciphertext it cannot open |
| What its AI can read | It has no built-in AI | Structure, never a secret value |
| Who holds the key | You | You |
| Starts at | $0 per member, per month | $0 |

Read from the vendor on 2026-08-02: https://obsidian.md/pricing

## Are Obsidian plugins safe?

Mostly, and Obsidian is straight with you about the limits. Restricted Mode is on by default, every plugin version is scanned, and popular ones get a manual review. But the documentation also states plainly that plugins "can access files on your computer, connect to the internet, and install additional programs", and that "due to technical limitations, Obsidian cannot reliably restrict plugins to specific permissions or access levels". Its own advice, for anyone with sensitive data, is to audit the plugin yourself first.

lodos took the other road, and gave up the ecosystem to do it. An extension here is a skill, a skill is instructions rather than code, and the app enforces which tools each one is allowed to call. There is nothing to audit because there is nothing to run.

## How an agent spends a key it cannot read

This is the part a notebook has no reason to have, so here it is concretely.

1. Your page references a credential the way it would a variable: `{{secrets.stripe.sk_live}}`. The reference is what gets stored, and the reference is all the model ever sees.
2. When the agent runs a command that needs the real value, that value is decrypted on your device and injected into an isolated subprocess as an environment variable.
3. The model gets the exit code and the output, with anything shaped like a secret stripped out. The value was never in its context, so it cannot surface in a transcript, a log, or an answer six weeks from now.

The audit log records that a secret was used, by whom and for what, and never which value. Entries are hash-linked, so removing one is detectable.

## A week in Obsidian, a week in lodos

Same company, same week, same three jobs.

### Last week’s numbers

- **Obsidian:** You open Stripe, read the figures, and type them into a note so that next week you have something to compare against.
- **lodos:** A workflow ran at 08:00. It queried Stripe with a key the model never saw, filed the digest to your wiki with its sources, and pinned the number to your dashboard.

### A customer asks for a refund

- **Obsidian:** Your note explains the procedure perfectly. You still open the dashboard, find the key in another app, and do every step by hand.
- **lodos:** You say what happened. The agent follows the refund playbook, spends the key inside an isolated subprocess, writes the outcome to the wiki and closes the task.

### Why did we choose this architecture?

- **Obsidian:** Search and the graph get you to nine pages quickly. Three are out of date, and only you can tell which three.
- **lodos:** The agent answers from the index with citations, and last night’s pass had already marked the out-of-date pages stale.

## Should you switch?

- Switch if the notes were never the whole job. If your week is credentials, follow-through and things that should have happened on a schedule, that is the part lodos was built for.
- Stay on Obsidian if you want a notebook, or if most of your writing happens on a phone. It is better at both and will be for a while.
- Or try it without deciding anything. lodos reads the vault where it sits and changes nothing in it, so running both for a fortnight costs you a download and no risk.

## Questions people actually ask

### Is there an Obsidian alternative with tasks and a password vault built in?

That is what lodos is. Same local-first markdown and the same graph, plus a kanban board, an encrypted vault, scheduled workflows and an agent that can act on all three. The trade is Obsidian’s plugin ecosystem and its mobile apps, neither of which lodos has today.

### Can lodos import my Obsidian vault?

Yes, and there is no export step because the files are already markdown. Pages, folder structure and [[links]] come across, each file also becomes an immutable hash-checked source, and anything holding a plaintext secret is flagged so it can move into the vault instead. What does not come across: attachments, images, .canvas files and your .obsidian config, so themes and plugin settings stay behind. Your original folder is left exactly as it was.

### Is Obsidian Sync end-to-end encrypted?

Yes. Obsidian states that Sync uses AES-256 end-to-end encryption and that they cannot read your data. On that specific question Obsidian and lodos give the same answer, which is why this page argues about scope rather than about privacy.

### What can lodos do that Obsidian plugins cannot?

Use a credential without ever revealing it. A plugin runs with full access to your vault and your machine, so anything it touches, it can read. In lodos the agent references a secret, the value is injected into an isolated subprocess, and no tool in the system returns a plaintext secret at all.

### Does lodos have a graph view like Obsidian?

Yes, local and global, resolved by the same engine that renders the links and feeds the AI. The difference is not the picture, it is that the agent can traverse the graph to build context across several pages.

### Is lodos free like Obsidian?

The app is free and stays free, with no account: vault, wiki, graph, tasks, workflows, skills, documents, charts and the built-in AI. Only encrypted cloud backup is paid, at $9 a month for the whole install. The AI runs on your own Claude account, so we never resell you tokens.

### Is there a lodos mobile app?

Yes. The desktop runs on macOS, Windows and Linux, and the phone companion reads the same encrypted snapshot behind Face ID: read-only by design, so you can check a credential or approve a run away from your desk. Writing still happens on the desktop, which is also where the agent, the workflows and the vault live.

## lodos vs Obsidian, feature by feature

The same rows we publish on the product page, with the Obsidian column pulled out beside ours. Anything marked limited explains itself on hover.

| | lodos | Obsidian |
| --- | --- | --- |
| Knowledge wiki | Yes | Yes |
| Obsidian-style graph | Yes | Yes |
| Tasks & kanban | Yes | No |
| Workflows & automation | Yes | No |
| Zero-knowledge secret vault | Yes | No |
| Built-in AI agent | Yes | Obsidian’s AI depends on plugins; lodos ships a built-in agent with your company’s full context. |
| Local-first, fully offline | Yes | Yes |
| Your whole company, one app | Yes | No |

If this page describes Obsidian incorrectly, that is a bug. Tell us and we will fix it.
