Why RyTask

The problem

Small teams that live with interruptions have a hard time proving where their week went. Work arrives from every direction — a Slack DM, a busy channel, an email thread, a ticket marked urgent. Each interruption feels small. Together they consume the days you had budgeted for planned work, and at the end of the week you have no credible, low-effort way to show that the gap between plan and reality was the cost of firefighting, not a lack of focus.

Existing trackers fail this in one of two ways. Some ignore time entirely — they track tasks beautifully but cannot tell you what anything cost. Others can tell you, but only if you pay: time tracking, Slack capture, and honest reporting are exactly the features that tend to sit behind paid tiers, even in tools whose core is open source. And the tools deep enough for engineers are usually hostile to everyone else on the team, so the "source of truth" fragments back into chat and spreadsheets.

The wedge

RyTask is built around three things that have to work together:

• Native time tracking with a planned-vs-interruption ledger. Every minute you track is either planned work or an interruption — there is no third bucket. That single constraint is what makes the weekly report honest: planned plus interruption always equals tracked, and the ledger of interruptions is the evidence behind the headline number. See the time-tracking model.
• Capture where the work arrives. Urgent work shows up in Slack, so RyTask captures it from Slack — a slash command or mention becomes a task in seconds, with the interruption recorded at the moment it happened rather than reconstructed later.
• Agents as first-class users. Anything a person can do in the UI, an AI agent can do through the built-in MCP server — and a CI gate fails the build if that parity ever slips. See MCP parity.

The nine differentiators

These are the commitments the product is steered by, each in one sentence.

No competitor combines all nine, and the most-needed pieces — Slack capture and time tracking — are exactly the ones that tend to be paywalled elsewhere.

The north-star metric

RyTask steers by one number: tasks captured-and-tracked per active user per week (CTW). A task counts only if it was both created with low friction (quick-add, Slack, or MCP) and had time logged against it.

The coupling is deliberate. Capture alone can be gamed — a tool can fill up with tasks nobody works on. Tracking alone proves nothing about whether the tool meets work where it arrives. A high CTW means both halves are true at once: capture is frictionless enough to record reality, and the tool is trusted enough that people log honest time against what they captured. That is the whole product in one metric.

The Albert and Marissa bar

Two personas act as a permanent product gate. Albert is a manager who wants status and accountability without learning query syntax or new vocabulary. Marissa is an ops teammate who creates and tracks her own work but will not tolerate a tool built for developers. Every surface in RyTask — every screen, every report, every notification — must be usable by both of them on first contact, with no training. A feature that fails this bar ships behind an advanced surface or not at all. It is the discipline most engineering-led tools never enforce, which is exactly why it is a differentiator.

Free because you run it

RyTask is licensed under AGPL-3.0 and designed to be self-hosted. That stance is also the business model's honest answer to paywalls: the features competitors gate — native time tracking today, unlimited automations when they land, single sign-on eventually — are free in RyTask because you provide the hardware. There is no metering to enforce and no tier to upsell inside the core product. If a paid offering ever exists, it covers organizational concerns and hosted convenience, never the capabilities the product exists to provide.

For what the license asks of you in return, see Contributing.