Mostafa Hosseini

Mostafa Hosseini

Simple software for real problems

About

I’m a marketer, entrepreneur, and software builder. I’ve been building digital products and helping businesses grow for 15+ years. Currently building Top10Q — a productivity app that helps business owners focus on what matters most — and WP Spell Check, a WordPress tool for finding and fixing website issues. I like simple products that solve real problems without adding more complexity.

Badges

Tastemaker
Tastemaker
Gone streaking 10
Gone streaking 10
Gone streaking
Gone streaking
Gone streaking 25
Gone streaking 25
View all badges

Forums

Do numbered marks + spoken notes beat one big screenshot for coding agents?

When I paste a screenshot into Claude Code or Cursor, a crop plus an arrow usually fixes the "where." What kept going wrong for me was the "why" once there are three or four things to fix on one screen.

What helped was talking while I draw. I circle a clipped label and say "let this wrap," then draw an arrow to a chart point and say "tooltip on hover and keyboard focus." Lasso puts [1] and [2] in the transcript at the exact words I said while making each mark. So the agent gets numbered marks with my intent attached, instead of a paragraph it has to match back to the image.

I also stopped pasting. A local coding agent on the same Mac can pull the latest annotated capture or voice brief itself through Lasso's local MCP server, and Lasso asks before the first agent call.

I wrote up the full workflow, with an example prompt for a UI fix, here: https://www.trylasso.online/guid...

Talanoa 1.3: connect Claude, ChatGPT and your AI apps to your inbox

1.3 lets your AI assistant work with your email, from your own computer.

More and more of my day happens in Claude. Whenever I needed an email there, I copied and pasted it by hand. So Talanoa now speaks MCP, the standard that AI apps use to reach your tools. Ask Claude "What did Marie say about the contract?" and it asks Talanoa.

•

4d ago

npm audit says my framework has a vulnerability. How do you decide whether it affects you?

My app runs on Next.js. During a merge, the build printed a notice that my version is affected by a known security vulnerability. It named no advisory, and I m not an engineer, so my first instinct was to run whatever upgrade command it suggested. I didn t. The notice came from build output, not from me, so I treated it as information and had Claude Code investigate read-only first.

What it found: I was on Next.js 16.4.2, and npm audit listed two moderate advisories, both fixed in 16.4.5 or later. Both are cache leaks between users in Next s newer caching system. One needs a use cache function called through unstable_cache() on a site that uses root params. The other needs nested use cache functions whose stream metadata arrives late. A repo-wide search for unstable_cache, use cache , cacheComponents, cacheLife, cacheTag and root params found nothing, and the app has no locale-style routes. So the practical risk was very low.

The fix was still small. The newest 16.4.x is 16.4.6, same major version and the same React requirement. In a throwaway copy, only two lines in package.json changed (next and eslint-config-next), plus the Next packages in the lockfile, and npm audit no longer listed next. The flagged total went from 33 packages to 32.

The weak spots. A search shows my code doesn t use the feature. It doesn t prove that no dependency does, and I m relying on the advisory text for what affected means. I compared the new version s dependency lists from the registry, not its code, so I checked the upgrade on a preview build before it goes live. And the 32 other flagged packages are mostly dev tools and a service-worker build package. Fixing them needs bigger version jumps, so I left them as a separate job.

View more