moinism

[flagged]

Chat Octopus: AI video creator and editor.

Made a video about PG's 'Billionaire' essay:

video: https://chatoctopus.com/share/d93d661a-0e5b-43f3-946e-a9dea0... full chat: https://chatoctopus.com/share/chat/8d044aa2-dc48-47a0-aa65-1...

Every AI video editor on the market assumes your footage is already labeled

Shameless plug: I'm the founder of Chat Octopus, an AI media assistant, and it actually 'looks' at the videos to understand them before creating a cut.

Chat Octopus: Claude code for videos

https://chatoctopus.com

and now its flagged :D

congrats on the launch! do you guys have anything planned to test chat agents directly in the ui? I have an agent, but no exposed api so can't really use your product even though I have a genuine need.

Wow! congrats on the launch guys. client-side rendering is incredible, really. I saw your product somewhere and have it as an open tab in my chrome for ~2 weeks :D

I also saw another YC company, Mosaic, doing something similar. But your approach of chat-based editing is a lot closer to what I'm building. Shameless plug: I'm also working on a chat-based media processor. https://chatoctopus.com

But you guys are way ahead! will be looking at you for inspiration.

https://ffmpeg-api.com

As the domain (hopefully) indicates, A REST API for the FFmpeg service. So far, it's been a plain API, but now it's adding MCPs and AI endpoints, so you don't have to remember ffmpeg commands.

Wait, the CLI command for the package is.. 'tit'?

This looks good, but a bit overkill for what I'm trying to build tbh.

Haven't tried it yet, but it looks similar to OpenAI's. What is your experience?

I agree, technically, "sub agent" is also another tool. But I think it's important to differentiate tools with deterministic input/output from those with reasoning ability. A simple 'Tool' will take the input and try to execute, but the 'subagent' might reason that the action is unnecessary and that the required output already exists in the shared context. Or it can ask a clarifying question from the main agent before using its tools.

Dude, it looks great, but I just spent half an hour learning about its 'CodeAgents' feature. Which essentially is 'actions written as code'.

This idea has been floating around in my head, but it wasn't refined enough to implement. It's so wild that what you're thinking of may have already been done by someone else on the internet.

https://huggingface.co/docs/smolagents/conceptual_guides/int...

For those who are wondering, it's kind of similar to the 'Code Mode' idea implemented by Cloudflare and now being explored by Anthropic; Write code to discover and call MCPs instead of stuffing context window with their definations.

This looks good. Even though it's only in Python, I think its worth a try. Thanks.

Thank you. But the main blocker for me right now is their skill definition: https://platform.claude.com/docs/en/agent-sdk/skills#how-ski...

How do you deal with the different APIs/Tooluse schema in a custom build? As other people have mentioned, it's a bigger undertaking than it sounds.

Amen. Been seeing these agent SDKs coming out left and right for a couple of years and thought it'd be a breeze to build an agent. Now I'm trying to build one for ~3 weeks, and I've tried three different SDKs and a couple of architectures.

Here's what I found:

- Claude Code SDK (now called Agent SDK) is amazing, but I think they are still in the process of decoupling it from the Claude Code, and that's why a few things are weird. e.g, You can define a subagent programmatically, but not skills. Skills have to be placed in the filesystem and then referenced in the plugin. Also, only Anthoripic models are supported :(

- OpenAI's SDK's tight coupling with their platform is a plus point. i.e, you get agents and tool-use traces by default in your dashboard. Which you can later use for evaluation, distillation, or fine-tuning. But: 2. They have agent handoffs (which works in some cases), but not subagents. You can use tools as subagents, though. 1. Not easy to use a third-party model provider. Their docs provide sample codes, but it's not as easy as that.

- Google Agent Kit doesn't provide any Typescript SDK yet. So didn't try.

- Mastra, even though it looks pretty sweet, spins up a server for your agent, which you can then use via REST API. umm.. why?

- SmythOS SDK is the one I'm currently testing because it provides flexibility in terms of choosing the model provider and defining your own architecture (handoffs or subagents, etc.). It has its quirks, but I think it'll work for now.

Question: If you don't mind sharing, what is your current architecture? Agent -> SubAgents -> SubSubAgents? Linear? or a Planner-Executor?

I'll write a detailed post about my learnings from architectures (fingers crossed) soon.

Hey, this is super cool. congrats on the product and the launch!

I'm building something exactly similar and couldn't believe my eyes when I saw the HN post. What i'm building (chatoctopus.com) is more like a chat-first agent for video editing, only at a prototype stage. But what you guys have achieved is insane. Wishing you lots of success.

to healthy competition!

Congrats on the launch! Having support for regional voices is going to open up so many opportunities.

It's a shameless plug, but I have built and use something similar: https://backupdiary.com

Here are a few things you can store:

- Important passwords

- Insurance policies

- Anything owed or lent

- Online subscriptions to cancel

- Bills or loans to pay to avoid late fees

- Any information that your business partners would want to know

- Contact information for other important people in your life, such as close friends

- Any accounts or assets that you have and the login information for them

- Special instructions or wishes

As another Pakistani, I don't understand the need for a protest... Let the names be used.

Congrats on the launch! Just tried the demo and it looks impressive. Good luck.

Are you by any chance hiring global-remote, full-stack/front-end devs? Would love to work with you guys.

can you please share one such example or a package in python, node, etc?

You're right!

I'll fix the privacy policy right away. Thank you for pointing it out.

The note data is encrypted at-rest. The encryption keys are kept separate. Which means even if the whole database gets leaked/hacked, the note details won't be readable.

I plan to pursue SOC 2 certification, to give users peace of mind regarding snooping, etc., if I get some signs of product-market-fit.

Note is decrypted and sent as plain text PDF file to the recipients.

That is a good suggestion and doable too.

But if the recipient forgets the password? They won't be able to reset it, in most case either, because the reset link goes to the original email which the recipient may not have access to.

I'm the solo founder. My name is Moin Uddin. More about me here:

https://moin.im

https://twitter.com/moinism

https://github.com/moinism

The note is stored in the DB after its encrypted. What do you suggest I change to make it more secure?

That is understandable.

The note is stored encrypted using AES 256-bit keys. The only time it's decrypted without your credentials is when you don't login during the check-in period.

I also want to make it end-to-end encrypted (still looking into it) but that has a UX problem. If the recipients need a key to decrypt it, a new problem of keeping the key safe and not forget about it also appears. And if we keep the key on our server then it also kills the whole point of e2e.

I'd love to know any change I can make to make your more confident in it?

It does sound risky. But I wanted it to be secure enough so I could (and I am) store my own sensitive info too.

All the information your write in your note is encrypted using 256-bit AES key. Encryption keys and the data itself are stored in different services. So even if the database is leaked/hacked, the information will not be public.