Didon
Summary
Most freelancers discover where their time actually went by accident — a Toggl export at month-end, a panicked invoice reconstruction, a vague memory that Tuesday felt productive but the deliverables say otherwise. Didon exists to close that gap automatically, without asking you to remember to start a timer.
Didon runs as a background process on Mac, takes periodic screenshots, and feeds them through a locally-hosted Qwen-3-VL:2b model that writes a structured work journal without any manual input. You define your projects and activity categories upfront; the AI slots every window, file, and task into that context as the day progresses. The resulting logs are queryable in plain language and exportable to CSV for billing or reporting. The local-only architecture means your screen data never leaves the machine — a meaningful distinction if you're working on client material under NDA. The single-device license and Mac-only availability are hard ceilings for anyone working across machines or on Windows.
Bottom line: Didon earns its place for a solo Mac-based freelancer who bills by the hour and has never trusted themselves to start a timer — but the moment you need multi-device sync, team dashboards, or Windows support, the architecture has no answer.
Pricing Plans
- Price
- €89 one-time (lifetime access)
Lifetime
One-time payment for lifetime access, updates, and support
- Unlimited CSV exports
- Local LLM (Qwen-3-VL:2b)
- Context-aware categorization
- Natural language journal queries
View full pricing on didon.app →
Pricing may have changed since last verified. Check the official site for current plans.
Community Performance Report Card
No community ratings yet. Be the first to rate this tool!
Community Benchmarks Community
Sign in to submit a benchmarkNo community benchmarks yet. Be the first to share a real-world data point.
Pros
Sign in to edit- All screen analysis runs locally via Qwen-3-VL:2b, so screenshots and work data never reach an external server — which means client confidentiality obligations do not conflict with using the tool.
- Context-aware categorization lets you define your own project names and activity types upfront, so the AI logs time against your actual work structure rather than generic app categories you have to remap later.
- Auto-launch and auto-pause tied to system sleep means tracking starts from the first minute and stops the moment you step away, so idle time does not inflate your reported hours.
- Plain-language queries against the work journal let you ask 'how much time did I spend on Acme this week' without building a report — which matters on a Friday when a client asks and you need an answer in two minutes.
- CSV export of daily logs is included without restriction, so billing reconstruction and end-of-month invoicing do not require a manual audit of your own memory.
Cons
Sign in to edit- The license covers installation on one machine only. If you work across a laptop and a desktop on the same day, one machine goes untracked — there is no sync or secondary-device option, and the only workaround is running separate instances under separate purchases.
- Windows support does not exist in the shipping version — only a waitlist. A team where even one member is on Windows cannot standardize on Didon, and those users have to maintain a separate tracking workflow entirely.
- The natural language query interface operates against the locally stored log files, but there is no described API or external integration path. Teams that need time data flowing into a project management tool, accounting system, or calendar will be copying CSV exports by hand until integrations ship — a task the vendor lists as upcoming but not yet available.
- The local LLM is fixed at Qwen-3-VL:2b. There is no described option to swap in a different model if analysis accuracy on your specific workflow proves inconsistent. Teams that try the tool and find categorization errors on niche technical work have no tuning lever beyond adjusting the context configuration.
Community Reviews
Sign in to write a reviewNo reviews yet. Be the first to share your experience.
About
- Platforms
- macOS
- API Available
- No
- Self-Hosted
- No
- Last Updated
- 2026-06-22T08:02:11.083Z
Best For
Who it's for
- Solo founders and freelancers needing private time tracking
- Mac users wanting local AI analysis of screen activity
- Teams or individuals tracking billable hours without manual logging
- Users seeking context-aware categorization of coding, marketing, and other tasks
What it does well
- Automatic workday time tracking
- Generating exportable daily work logs for billing or reporting
- Analyzing time spent on projects and activities via natural language queries
- Improving productivity habits through AI insights on work patterns
Discussion Community
Sign in to commentNo discussion yet. Sign in to start the conversation.
Compare Didon
Spotted incorrect or missing data? Join our community of contributors.
Sign Up to ContributeCommunity Notes & Tips Community
Sign in to contributeBe the first to contribute. General notes, observations, gotchas, and tips from people who use this tool day-to-day.
Frequently Asked Questions
- Is Didon free?
- Didon is a paid tool (€89 one-time (lifetime access)). No permanent free tier is offered.
- Is Didon open source?
- No — Didon is a closed-source tool. Source code is not publicly available.
- What platforms does Didon support?
- Didon is available on: macOS.
Hours Saved & ROI Stories Community
Sign in to contributeBe the first to contribute. Concrete time/cost savings, with context. e.g. "Cut my code review backlog from 4h to 45m per week."
Curated lists that include this category
Keeping an accurate record of where a workday went is a problem that manual timers solve in theory and nobody maintains in practice. Didon addresses this by running silently in the background, capturing periodic screenshots, and using a local large language model to convert what it sees into a timestamped work journal. You configure the tool once with your project names and activity categories — coding, marketing, content creation, whatever your work actually contains — and from that point the tracking is automatic. Logs are saved on-device, queryable in plain language, and exportable to CSV for client invoices or internal reporting.
The distinguishing architectural choice is that all screen analysis runs on-device using Qwen-3-VL:2b, a local vision-language model. The vendor states that screenshots and sensitive data never leave the Mac. For consultants, lawyers, or developers working in client environments with data-handling obligations, this is not a minor footnote — it is the reason to choose this tool over a cloud-based tracker. There are no external API costs because no external API is called.
Didon fits a specific profile well: a solo operator on a single Mac who wants passive, accurate time data without behavioral change. The auto-launch and auto-pause logic — starting with the system, pausing when the Mac sleeps — means idle time is not counted and no manual intervention is required. Where it breaks is equally specific: the license covers one machine, there is no Windows client at launch, there are no team features, and the upcoming integrations and advanced reports listed on the vendor page are not yet available. A freelancer who switches between a laptop and a desktop loses coverage on one of them. A small agency wanting shared project visibility has no path forward in the current version.