PrepDojo 🥋

An interview prep and job application tracker that lives in your Obsidian vault.

Job hunting means doing a lot of small things every day. A LeetCode problem at breakfast, some ML review at night, a batch of applications in between. None of it feels like much, and a week later you can't remember any of it. PrepDojo keeps all of it on one dashboard page. You log with a click, or just tell an AI "did two sum today", and the same page shows your streak, which topics need work, how many applications went out this week, and which resume is actually getting you interviews.

Everything lives in plain files in an Obsidian vault on your own computer and your records stay readable in any editor.

✨ What you can do

⟳ Import your LeetCode solves with one click. Just crushed a few problems on LeetCode? Click ⟳ Import on the dashboard and your recent accepted submissions are logged for you.

＋ Log any prep work with one click. An application fired off, an ML concept reviewed at lunch, a mock interview survived: every kind of prep has its own button. Click, answer a prompt or two, logged.

✎ Track every application to the end. Heard back? Hit ✎ Update and log the new status in seconds.

⇄ A/B test your resume. Every application you send tests one resume version. The dashboard computes an interview rate per version, so you can see which one actually gets you interviews.

▦ One page shows it all. Everything above lives on a single dashboard: your current streak, topics ranked from solid to shaky, interviews and offers up top, resume rates below.

✦ Or let AI do the logging. Too tired to click? Toss Claude a LeetCode link or mumble "did two sum today" and the entry writes itself. Applications too: "Stripe moved me to phone screen" updates your tracker.

📋 Requirements

• Obsidian
• Python 3.9+ for the one-command setup. No Python at all? See manual setup.

Optional: the Claude desktop app (Cowork) or Claude Code, if you want AI logging.

🚀 Quick start

Step 1: Get the repo

git clone https://github.com/aaaxy/PrepDojo.git
cd PrepDojo

Step 2: Prepare your vault

1. Pick the folder that is (or will be) your Obsidian vault. It should live outside this repo.
2. Open that vault in Obsidian at least once.
3. On Obsidian 1.12+: open Settings → General → enable Command line interface.
4. Quit Obsidian (setup reopens it when installing the plugins).

Step 3: Run setup with your vault's path, back in the terminal:

python3 setup.py /path/to/YourVault

Setup asks two optional questions along the way: your LeetCode username (powers the import button) and your daily-notes folder. Everything else has sensible defaults you can change later. See docs/configuration.md.

Step 3b: skip to Step 4 if your Obsidian is 1.12 (released February 2026) or newer. On older versions, setup can't install the plugins, so install them by hand:

1. Open Settings → Community plugins and turn off Restricted mode.
2. Click Browse, then install and enable Dataview, QuickAdd, and Calendar.

Step 4: Finish in Obsidian, whichever version you're on:

1. Open Settings → Community plugins and open Dataview's settings (the ⚙️ next to it, or the ··· menu on newer versions). Turn on Enable JavaScript queries.
2. Open the dashboard note in Reading view and click a button. Done.

The tables start empty. Your first log fills them, and your streak starts counting.

Step 5 (optional): Add AI logging. Install once, in Claude Code or the Claude app:

/plugin marketplace add aaaxy/PrepDojo
/plugin install prepdojo@prepdojo

Your settings travel with your vault (setup writes a small prepdojo.json there), so the plugin needs no configuration and keeps working when you change settings.

That's it. You're all set. From here on, PrepDojo is just the dashboard and your daily grind.

---

Everything below is for later: changing settings, regenerating after a change, and updating PrepDojo itself.

🔄 Regenerating and updating

The dashboard, templates, and logging flows are all generated from config.toml. The same three steps apply whether you changed a setting or are updating PrepDojo itself:

1. Quit Obsidian.
2. In the repo (after git pull, if updating PrepDojo itself), run:

python3 generate.py /path/to/YourVault

The path is remembered in config.toml, so a plain python3 generate.py works from then on.

3. Restart Obsidian and reopen the dashboard note.

Reading the output: installed and up to date mean done. PRESERVED ... new version written to *.new means you've customized that file since it was installed. Compare it with the .new copy and merge at your own pace, or rerun with --force to take the new version wholesale.

What it can touch: generated files you haven't customized. Those update in place.

What it never touches: anything you have customized (kept as is, with the fresh version written beside it as .new) and your data. Daily notes and the application CSVs are never written, under any flag.

Note

Because your CSVs are never touched, an update can't add a new column to them either. When a release adds one to the application tracker, the release notes will ask you to add that column name to the end of your applications.csv header row yourself.

The AI plugin updates separately: it updates through Claude's plugin manager, not generate.py. After a PrepDojo release, refresh it from the /plugin menu in Claude.

🔍 System overview

To you, PrepDojo is one dashboard page. Under the hood, that page is built from three layers, all generated from one config file:

• Capture: log what you did in seconds, from wherever you are. Four ways into the same log: one-click dashboard buttons with guided prompts, QuickAdd hotkeys running the same flows from anywhere in Obsidian, an AI plugin for Claude, and a CLI (prepdojo.py) for terminals and automation. The AI plugin turns "just did two sum" or a pasted LeetCode URL into a correctly formatted entry. Its skills read plain prepdojo.json, so any LLM agent can follow the same contract. A fifth way is automatic: the ⟳ LeetCode import fetches your recent accepted submissions. Every entry point validates entries, so the log can't drift into inconsistency
• Storage: where your record lives. Plain text files on your own machine: markdown bullets in your daily notes for practice, CSV rows for job applications. Human-readable, grep-able, no database, no lock-in. Every tool in this repo is replaceable, and your record outlives all of them
• Insight: how you see your progress. One dashboard note computes it all from your record: streaks, per-topic and per-difficulty breakdowns with date-range filters, a "needs re-review" list, and application stats (active interviews, offers, milestone summary, interview rate per resume version), updating live while the note is open. It stores nothing itself, so regenerating, moving, or customizing it is always safe

Daily workflow, from a solved problem to insight:

flowchart LR
    subgraph capture ["⚡ Capture (seconds)"]
        subgraph obs ["on the dashboard"]
            BT["＋ / ✎ buttons<br/>guided prompts + dropdowns"]
            IMP["⟳ LeetCode import<br/>recent solves, one click"]
        end
        subgraph obs2 ["anywhere in Obsidian"]
            QA["QuickAdd hotkeys<br/>same flows as the buttons"]
            MB["Manual bullet<br/>type it yourself"]
        end
        subgraph cl ["chat with Claude"]
            CS["AI plugin<br/>paste a link or say 'log it'"]
        end
        subgraph term ["terminal / scripts"]
            CLI["prepdojo CLI<br/>hooks, automation, sync-lc"]
        end
    end
    subgraph storage ["📝 Storage (plain files)"]
        DN["Daily notes<br/><code>- LC #200 Number of Islands · Medium · bfs/dfs #lc</code>"]
        CSV["applications.csv<br/>resume-versions.csv"]
    end
    subgraph insight ["📊 Insight (always current)"]
        DB["Dataview dashboard<br/>streaks · topics · interviews & offers · resume rates"]
    end
    BT --> DN
    IMP --> DN
    QA --> DN
    MB --> DN
    CS --> DN
    CLI --> DN
    BT --> CSV
    CS --> CSV
    CLI --> CSV
    DN --> DB
    CSV --> DB

Loading

One-time setup, everything personalized from a single file:

flowchart LR
    CFG["config.toml<br/>your few answers<br/>+ built-in defaults"] --> GEN["generate.py"]
    GEN --> T["Daily note template"]
    GEN --> D["Dashboard note"]
    GEN --> Q["QuickAdd actions<br/>+ logging scripts"]
    GEN --> CSVS["Starter CSVs<br/>(created once, never overwritten)"]
    GEN --> PJ["prepdojo.json<br/>(read by the AI plugin)"]
    T & D & Q & CSVS & PJ --> V["Your Obsidian vault"]

Loading

📜 License

PolyForm Noncommercial 1.0.0: free to use, modify, and share for any noncommercial purpose (personal use, education, research, nonprofits). Commercial use requires a separate license from the author.

Copyright (c) 2026 aaaxy