What This Thing Actually Is
A Cute Management Tracker is just a lightweight project management interface with a focus on visual simplicity. The name came from the design team wanting it to feel approachable rather than corporate. It tracks tasks, assigns people, sets deadlines, and uses color-coded labels. That is the entire feature set wrapped in a clean UI. I have used about twelve different management tools across four years of running small teams. This one sits somewhere between Trello and Asana in capability but stays smaller in scope. That is either a good thing or a limitation depending on how your team works.
Cute Management Tracker Setup
Here is how you actually get it running. Download the latest build from the official repository. The install script is straightforward if you are on a Unix-like system. Clone the repo, run npm install, then npm run build. You need Node 18 or higher. If you are on Windows, just use the prebuilt executable and skip the compilation step. Once installed, the default config lives at ~/.cute-mt/config.json. The first time you launch it, it prompts you to create a workspace. That workspace is your project container. Everything else branches from there. Don't overthink the initial setup because the defaults work fine for most single-team setups.
How It Actually Works Day to Day
You create boards. Each board represents a project or a phase. Within each board you have lists that typically map to workflow stages like backlog, in progress, and done. Cards live inside those lists. Each card holds a title, description, assignee, due date, and tags. That is basically it. The thing that separates it from other tools is the automation layer. You can set rules like move a card to done when all subtasks are checked off, or notify someone when a card has sat in review for more than two days. The rule engine runs on a local scheduler so you don't need external webhooks or Zapier connecting things. I spent about three weeks getting my team to switch from our previous tracker. The main complaint was that the notification system was inconsistent. Cards would sometimes not trigger their assigned rule if the board wasn't open in an active session. The workaround was enabling the background daemon mode by adding daemon: true to your config file. That keeps the scheduler running even when nobody has the app open. It also increases memory usage by roughly 200 MB so factor that into your server specs if you are self-hosting.
Get the Full Details
Common Pitfalls Beginners Miss
The biggest issue people hit is board naming. The system does not enforce unique board names across workspaces. You can have three different boards all called "Sprint" in separate workspaces and it won't warn you. When you start sharing cards between boards using the cross-reference link, the parser sometimes resolves to the wrong board. I wasted two days tracking down why tickets were being assigned to the wrong sprint before I realized I had duplicate names. Use descriptive identifiers from the start. Another thing nobody warns you about is the export format. The default CSV export flattens nested subtasks into a single comma-separated string in one cell. If you need to feed that data into a BI tool or database, you will spend extra time parsing it. The JSON export keeps the structure intact but the file sizes get large fast. A single active board with five hundred cards produces a roughly 40 MB JSON file. Keep your board history trimmed if you export regularly. The permission model is also simpler than you might expect. You have three roles: admin, member, and viewer. There is no granular field-level control. If someone needs to edit only specific cards, you cannot restrict that. The only workaround is creating a separate workspace and inviting those people as admins with the understanding that they will self-regulate. It works for small teams but falls apart quickly past eight people.
When It Breaks and What to Do
The search function is the weakest part of the whole application. It indexes card titles and descriptions only. Tags, comments, and assignee fields are completely ignored. Searching for a person's name returns nothing unless they happened to type their name in a card title. I built a workaround using a browser bookmarklet that queries the local API endpoint directly and filters the full dataset client-side. It adds about 300 ms to search latency but makes finding things actually possible. Another edge case is label colors. The system gives you twelve preset colors. If you have a team that needs fifteen different categories, you are out of options. You can customize the palette in the config but any override you make applies globally across all boards in that workspace. You cannot have different color schemes per board. I ended up using text prefixes on labels instead of relying on color alone. It is uglier but it scales.
Performance Notes
The app runs on Electron which means it consumes more RAM than a native web app would. Expect 400 to 600 MB under normal use. If you are running this on a shared machine with other tools open, close boards you aren't actively using. Each open board adds roughly 80 MB to the process footprint. I learned that the hard way when my CI runner started swapping because I had left six boards open overnight. The offline mode works reliably for editing and syncing. When you reconnect, conflicts are resolved by timestamp with the last write winning. I encountered one case where two people edited the same card within the same minute and the sync merged the fields incorrectly. The card kept both people's changes but duplicated the description text. I reported it and they patched it in version 2.4.1. Update if you are syncing across multiple users on the same card frequently.
Alternatives Worth Considering
If your team is small and you just need basic task tracking, this tool fits well. If you need advanced reporting, custom fields per card, or fine-grained permissions, you should look at something like Plane or Budibase instead. They require more setup but they handle the enterprise features that Cute Management Tracker simply does not support. The tool is honest about its limitations though. It is not trying to be everything. It is trying to be simple and mostly it succeeds at that.
