DoneThat vs Sidekick
Open-source assistant that can reference on‑device files and webpages.
AI Memory: side-by-side
Tools to archive your digital life and make it available for later use by eg AIs.
| Tool | Input data | Platform support | Data location | API |
|---|---|---|---|---|
Privacy-first work tracker and AI coach Best for: Calendar-like timeline of your work. | Screen, audio, apps | macOS, Windows, Linux | Cloud | Yes |
Open-source assistant that can reference on‑device files and webpages. Best for: Mac users wanting local AI memory over personal data. | Local files/web | macOS | Local | -No |
Sidekick is best for
Mac users wanting local AI memory over personal data.
DoneThat is best for
Calendar-like timeline of your work.
See the full ai memory comparison or visit Sidekick.
AI-generated comparison
Prompt
Compare https://donethat.ai/ and https://github.com/johnbean393/Sidekick honestly. Check both websites yourself, then summarize where DoneThat is stronger and where Sidekick is stronger. Do not rank them globally.
Result
DoneThat and Sidekick solve different problems: DoneThat is a cross-platform automated time-tracking and productivity coaching app with a paid subscription and team features, while Sidekick is a free, open-source macOS-native local LLM chat app focused on private on-device inference and document/web research.
DoneThat stronger
- Runs on Windows, Mac, and Linux, whereas Sidekick is restricted to Apple Silicon Macs with 8GB+ RAM.
- Purpose-built for passive time tracking and timesheet automation, which Sidekick does not attempt to offer.
- Ships an AI productivity coach ('Don') that interprets your actual work activity for goal-setting, rather than a general chat interface.
- Provides team-oriented capabilities like automated timesheets, meeting reduction, and workflow integrations (API, Zapier, n8n).
- Offers tiered paid plans with priority support and a 14-day trial, which is a clearer commercial path for buyers who want vendor accountability.
Sidekick stronger
- Free and open-source under MIT with no subscription, which is hard to beat for individual users on a budget.
- Runs LLM inference fully locally via llama.cpp with GGUF models, keeping conversations offline for maximum privacy.
- Includes RAG over local files, folders, and websites through configurable 'experts', which DoneThat does not advertise.
- Bundles a broader AI toolkit: Deep Research agent across 50-80 webpages, Canvas for code/content editing, image generation, and function calling.
- Supports bring-your-own-key for many OpenAI-compatible providers (OpenAI, Anthropic, Google, DeepSeek, Groq), giving power users flexible model choice without a vendor's pricing tiers.
Explore related pages
Productivity Coaching
Meet Don, the AI coach inside DoneThat that turns your work patterns into practical coaching, goals, and next steps.
ExploreFeature Overview
Explore DoneThat features for AI time tracking, automatic summaries, productivity coaching, privacy controls, and team visibility.
ExplorePrivacy
Learn how DoneThat protects your data with privacy controls, secure architecture, local capture options, and AI data handling.
ExploreReady to try DoneThat?
Fully automated AI time tracking and coaching. 14-day free trial.