Best new AI, design and productivity resources: day 34

Yesterday's additions span practical ways to organize work, inspect what AI agents actually do, and create better interfaces. These are curated starting points, not endorsements of every claim made by the projects. Follow the original links to check whether each fits your workflow.

Find ideas and keep projects moving

What Ships collects startup launch videos from X, with links back to the original posts. If you are preparing a product demo, browse examples to see how other teams introduce a product rather than starting with a blank page. https://whatships.com/

TaskView is a self-hosted project and task management option for teams who want direct control over their workspace. It is a more relevant pick for organizing ongoing work than for one-off personal to-dos. https://github.com/Gimanh/taskview-community

Mulmoclaude takes a local-first approach to an AI assistant, keeping memories and small personal apps in files on your computer. Consider it if ownership of your working context matters more to you than an entirely hosted assistant. https://github.com/receptron/mulmoclaude

Make AI-agent work easier to inspect

Flyto Core focuses on recorded, replayable browser automation so you can inspect evidence of what an agent did instead of relying only on its final message. Test any automation on low-risk work before giving it access to consequential tasks. https://github.com/flytohub/flyto-core

Django Orbit brings agent-readable traces and debugging to Django applications. It is a focused choice if your team already builds with Django, rather than a general-purpose monitoring tool. https://github.com/astro-stack/django-orbit

Threadport packages portable, verifiable work state for coding agents; Agentlas OS organizes specialist agents around tasks. The first helps with continuity between sessions, while the second is aimed at coordinating multiple agents. https://github.com/Frankie-Xu/threadport

Agentlas OS: https://github.com/agentlas-ai/Agentlas-OS

Railtracks and Dive offer different ways to build agent systems: Railtracks is a framework for resilient agent workflows, while Dive is for writing agents in Go. Choose based on your stack and test the behavior in your own environment. https://github.com/RailtownAI/railtracks

Dive: https://github.com/deepnoodle-ai/dive

Build more consistent interfaces

Balsa UI is aimed at design systems that AI agents can read, from tokens to components. Springs Studio offers agent-readable easing and motion tokens. Together they address two common sources of inconsistency in generated interfaces: structure and movement. https://balsa-ui.com/

Springs Studio: https://www.springs.studio/

Meshfont is a free browser-based font design playground, and Best Free Fonts is a collection for finding existing typefaces. Use the former when you want to experiment with letterforms; use the latter when you need to choose a font quickly. https://meshfont.com/

Best Free Fonts: https://bestfreefonts.com/

ForgeUI and 21st offer interface components and templates, while Paper provides a place to design and share work. Check the code, licensing and compatibility of a component before using it in a production project. https://forgeui.in/

21st: https://21st.dev/

Paper: https://paper.design/

For more visual motion, Theatre.js provides an animation editor; Unicorn Studio offers interactive graphics; and Spline supports collaborative 3D design in the browser. Each has a different role, so pick the one that matches the medium you are creating. https://www.theatrejs.com/

Unicorn Studio: https://www.unicorn.studio/

Spline: https://spline.design/

More focused additions

CoSAI Risk Map provides a framework for thinking through AI-system security risks. Agent Rules Skill helps generate repository instructions for coding agents, while AI-First Toolkit collects skills for Claude Code and Codex. Review generated instructions and third-party skills before granting access to your code or accounts. https://github.com/cosai-oasis/secure-ai-tooling

Agent Rules Skill: https://github.com/netresearch/agent-rules-skill

AI-First Toolkit: https://github.com/techwolf-ai/ai-first-toolkit

Other additions include LiveContext CE for self-hosted AI workflows, People Context for local-first personal memory, OpenExecutive for virtual executive-team workflows, and Kiln for agent-assisted 3D printing. The complete directory has all of yesterday's additions and their original links. https://bestproductivityresources.com/directory