The best Hacker News stories from Show from the past day

from

All Show

from the past*

day week month

with these search terms:

Create follow Go back

Latest posts:

Show HN: Prism – Let browser agents access any app

Hey HN, We’re Alex, Land, and Rajit. We’re building Prism (prismai.sh), a tool that helps browser agents authenticate onto websites with user credentials. Developers pass in credentials, Prism logs into a website on their behalf, and hands them back the cookies so they have an authenticated session. Here’s an example of how developers can use Prism to complete username/password flows (<a href="https://youtu.be/SEtVUnWnxuE" rel="nofollow">https://youtu.be/SEtVUnWnxuE</a>), and here’s an example of how developers can use Prism to complete login flows that require an OTP code (<a href="https://youtu.be/fe9w9PvrwH0" rel="nofollow">https://youtu.be/fe9w9PvrwH0</a>).<p>We spoke to browser agent developers and saw people copying and pasting credentials and even credit card numbers directly into model system prompts. We were surprised that there wasn’t a better way to give agents access to websites on a human’s behalf. Moreover, we noticed that every company had to build infrastructure to manage OTP, TOTP, and MFA and that auth remained a significant hurdle in agent reliability. We wondered if this was a boring part of the problem of building web automations that someone could automate away.<p>We started working with Casco, an autonomous security testing company, to enable their agent to access customer sites. Before a pentest, Casco makes a request to Prism’s API specifying test user credentials, a domain, and a login method. For example, give me an authenticated session for the account rajit@prismai.sh for OpenAI via OTP code over email. Our agent logs in on their behalf (without exposing credentials to a model), and we download the cookies and send them back in the response.<p>To maintain speed and reliability, we use playwright in most cases to login (which gives us speed), and we fallback to AI on failure (which gives us reliability). We have a number of websites we support out of the box and add new scripts as the number of websites we need to support grows. We are working on a way for the agent to update the existing playwright script on failure, so our scripts always stay up to date.<p>To try our api, you can use our API playground docs.prismai.sh/api-reference/endpoint/login to sign into x.com with the following API key: pk_54abb1cd0a637eb973ed690416e71a953e98f2ea839cf16529bbfa41a41bc016 .<p>We’d love to learn more about how other developers give agents access to their accounts. We look forward to everyone’s feedback and comments.

Show HN: Prism – Let browser agents access any app

Hey HN, We’re Alex, Land, and Rajit. We’re building Prism (prismai.sh), a tool that helps browser agents authenticate onto websites with user credentials. Developers pass in credentials, Prism logs into a website on their behalf, and hands them back the cookies so they have an authenticated session. Here’s an example of how developers can use Prism to complete username/password flows (<a href="https://youtu.be/SEtVUnWnxuE" rel="nofollow">https://youtu.be/SEtVUnWnxuE</a>), and here’s an example of how developers can use Prism to complete login flows that require an OTP code (<a href="https://youtu.be/fe9w9PvrwH0" rel="nofollow">https://youtu.be/fe9w9PvrwH0</a>).<p>We spoke to browser agent developers and saw people copying and pasting credentials and even credit card numbers directly into model system prompts. We were surprised that there wasn’t a better way to give agents access to websites on a human’s behalf. Moreover, we noticed that every company had to build infrastructure to manage OTP, TOTP, and MFA and that auth remained a significant hurdle in agent reliability. We wondered if this was a boring part of the problem of building web automations that someone could automate away.<p>We started working with Casco, an autonomous security testing company, to enable their agent to access customer sites. Before a pentest, Casco makes a request to Prism’s API specifying test user credentials, a domain, and a login method. For example, give me an authenticated session for the account rajit@prismai.sh for OpenAI via OTP code over email. Our agent logs in on their behalf (without exposing credentials to a model), and we download the cookies and send them back in the response.<p>To maintain speed and reliability, we use playwright in most cases to login (which gives us speed), and we fallback to AI on failure (which gives us reliability). We have a number of websites we support out of the box and add new scripts as the number of websites we need to support grows. We are working on a way for the agent to update the existing playwright script on failure, so our scripts always stay up to date.<p>To try our api, you can use our API playground docs.prismai.sh/api-reference/endpoint/login to sign into x.com with the following API key: pk_54abb1cd0a637eb973ed690416e71a953e98f2ea839cf16529bbfa41a41bc016 .<p>We’d love to learn more about how other developers give agents access to their accounts. We look forward to everyone’s feedback and comments.

Show HN: Open-source AI data generator (now hosted)

Hey HN! A few months ago we shared our AI dataset generator as an open source repo, and the response was incredible (<a href="https://news.ycombinator.com/item?id=44388093">https://news.ycombinator.com/item?id=44388093</a>). We got requests from folks who wanted to use it without the hosting overhead, so we created both options: a hosted version (<a href="https://www.metabase.com/ai-data-generator" rel="nofollow">https://www.metabase.com/ai-data-generator</a> for instant use and the source code fully open (<a href="https://github.com/metabase/dataset-generator" rel="nofollow">https://github.com/metabase/dataset-generator</a>) for anyone who wants to self-host or contribute.<p>Looking forward to seeing how you use it and what you build on top of it!<p>Bonus: The repo now supports multi-provider LLM integration with LiteLLM, thanks to a great contribution from their team.

Show HN: Vibe Linking

Show HN: Vibe Linking

Show HN: Vibe Linking

Show HN: Vibe Linking

Show HN: Vibe Linking

Show HN: Vibe Linking

A vibrator helped me debug a motorcycle brake light system

Show HN: Dayflow – A git log for your day

Hi HN! I've been building Dayflow, a macOS app that automatically tracks what you're actually working on (not just which apps you have open).<p>Here's what it does:<p>- It creates a semantic timeline of your day;<p>- It does it by understanding the content on your screen (with local or cloud VLMs);<p>- This allows you to see exactly where your time went without any manual logging.<p>Traditional time trackers tell you "3 hours in Chrome" which is not very helpful. Dayflow actually understands if you're reading documentation, debugging code, or scrolling HN. Instead of "Chrome: 3 hours", you get "Reviewed PR comments: 45min", "Read HN thread about Rust: 20min", "Debugged auth flow: 1.5hr".<p>I was an early Rewind user but rarely used the retrieval feature. I built Dayflow because I saw other interesting uses for screen data. I find that it helps me stay on track while working - I check it every few hours and make sure I’m spending my time the way I intended - if I’m not, I try to course correct.<p>Here’s what you need to know about privacy:<p>- Run 100% locally using qwen2.5-vl-3b (~4GB model)<p>- No cloud uploads, no account<p>- Full source available under MIT license (<a href="https://github.com/JerryZLiu/Dayflow" rel="nofollow">https://github.com/JerryZLiu/Dayflow</a>)<p>- Optional: BYO Gemini API key for better quality (stored in Keychain, with free-tier workaround to prevent training on your data)<p>The tech stack is pretty simple, SwiftUI with a local sqlite DB. Uses native macOS apis for efficient screen captures. Since most people who run LLMs locally already have their tool of choice (Ollama, LLMStudio, etc.), I decided to not embed an LLM into Dayflow.<p>By far the biggest challenge was adapting from SOTA vision models like Gemini 2.5 Pro to small, local models. My constraints were that it had to take up <4GB of ram and have vision capabilities. I had to do a lot of evals to figure out that Qwen2.5VL-3B was the best balance of size and quality, but there was still a sizable tradeoff in quality that I had to accept. I also got creative with sampling rates and prompt chunking to deal with the 100x smaller context window. Processing a 15 minute segment takes ~32 local LLM calls vs 2 Gemini calls!<p>Here’s what I’m working on next:<p>Distillation: Using Gemini's high-quality outputs as training data to teach a local model the patterns it needs, hopefully closing the quality gap.<p>Custom dashboards where you can track answers to any question like "How long did I spend on HN?" or "Hours until my first deep work session of the day<p>I'd love to hear your thoughts, especially if you've struggled with productivity tracking or have ideas for what you'd want from a tool like this.

Show HN: Dayflow – A git log for your day

Hi HN! I've been building Dayflow, a macOS app that automatically tracks what you're actually working on (not just which apps you have open).<p>Here's what it does:<p>- It creates a semantic timeline of your day;<p>- It does it by understanding the content on your screen (with local or cloud VLMs);<p>- This allows you to see exactly where your time went without any manual logging.<p>Traditional time trackers tell you "3 hours in Chrome" which is not very helpful. Dayflow actually understands if you're reading documentation, debugging code, or scrolling HN. Instead of "Chrome: 3 hours", you get "Reviewed PR comments: 45min", "Read HN thread about Rust: 20min", "Debugged auth flow: 1.5hr".<p>I was an early Rewind user but rarely used the retrieval feature. I built Dayflow because I saw other interesting uses for screen data. I find that it helps me stay on track while working - I check it every few hours and make sure I’m spending my time the way I intended - if I’m not, I try to course correct.<p>Here’s what you need to know about privacy:<p>- Run 100% locally using qwen2.5-vl-3b (~4GB model)<p>- No cloud uploads, no account<p>- Full source available under MIT license (<a href="https://github.com/JerryZLiu/Dayflow" rel="nofollow">https://github.com/JerryZLiu/Dayflow</a>)<p>- Optional: BYO Gemini API key for better quality (stored in Keychain, with free-tier workaround to prevent training on your data)<p>The tech stack is pretty simple, SwiftUI with a local sqlite DB. Uses native macOS apis for efficient screen captures. Since most people who run LLMs locally already have their tool of choice (Ollama, LLMStudio, etc.), I decided to not embed an LLM into Dayflow.<p>By far the biggest challenge was adapting from SOTA vision models like Gemini 2.5 Pro to small, local models. My constraints were that it had to take up <4GB of ram and have vision capabilities. I had to do a lot of evals to figure out that Qwen2.5VL-3B was the best balance of size and quality, but there was still a sizable tradeoff in quality that I had to accept. I also got creative with sampling rates and prompt chunking to deal with the 100x smaller context window. Processing a 15 minute segment takes ~32 local LLM calls vs 2 Gemini calls!<p>Here’s what I’m working on next:<p>Distillation: Using Gemini's high-quality outputs as training data to teach a local model the patterns it needs, hopefully closing the quality gap.<p>Custom dashboards where you can track answers to any question like "How long did I spend on HN?" or "Hours until my first deep work session of the day<p>I'd love to hear your thoughts, especially if you've struggled with productivity tracking or have ideas for what you'd want from a tool like this.

Show HN: Dayflow – A git log for your day

Hi HN! I've been building Dayflow, a macOS app that automatically tracks what you're actually working on (not just which apps you have open).<p>Here's what it does:<p>- It creates a semantic timeline of your day;<p>- It does it by understanding the content on your screen (with local or cloud VLMs);<p>- This allows you to see exactly where your time went without any manual logging.<p>Traditional time trackers tell you "3 hours in Chrome" which is not very helpful. Dayflow actually understands if you're reading documentation, debugging code, or scrolling HN. Instead of "Chrome: 3 hours", you get "Reviewed PR comments: 45min", "Read HN thread about Rust: 20min", "Debugged auth flow: 1.5hr".<p>I was an early Rewind user but rarely used the retrieval feature. I built Dayflow because I saw other interesting uses for screen data. I find that it helps me stay on track while working - I check it every few hours and make sure I’m spending my time the way I intended - if I’m not, I try to course correct.<p>Here’s what you need to know about privacy:<p>- Run 100% locally using qwen2.5-vl-3b (~4GB model)<p>- No cloud uploads, no account<p>- Full source available under MIT license (<a href="https://github.com/JerryZLiu/Dayflow" rel="nofollow">https://github.com/JerryZLiu/Dayflow</a>)<p>- Optional: BYO Gemini API key for better quality (stored in Keychain, with free-tier workaround to prevent training on your data)<p>The tech stack is pretty simple, SwiftUI with a local sqlite DB. Uses native macOS apis for efficient screen captures. Since most people who run LLMs locally already have their tool of choice (Ollama, LLMStudio, etc.), I decided to not embed an LLM into Dayflow.<p>By far the biggest challenge was adapting from SOTA vision models like Gemini 2.5 Pro to small, local models. My constraints were that it had to take up <4GB of ram and have vision capabilities. I had to do a lot of evals to figure out that Qwen2.5VL-3B was the best balance of size and quality, but there was still a sizable tradeoff in quality that I had to accept. I also got creative with sampling rates and prompt chunking to deal with the 100x smaller context window. Processing a 15 minute segment takes ~32 local LLM calls vs 2 Gemini calls!<p>Here’s what I’m working on next:<p>Distillation: Using Gemini's high-quality outputs as training data to teach a local model the patterns it needs, hopefully closing the quality gap.<p>Custom dashboards where you can track answers to any question like "How long did I spend on HN?" or "Hours until my first deep work session of the day<p>I'd love to hear your thoughts, especially if you've struggled with productivity tracking or have ideas for what you'd want from a tool like this.

Show HN: Dayflow – A git log for your day

Hi HN! I've been building Dayflow, a macOS app that automatically tracks what you're actually working on (not just which apps you have open).<p>Here's what it does:<p>- It creates a semantic timeline of your day;<p>- It does it by understanding the content on your screen (with local or cloud VLMs);<p>- This allows you to see exactly where your time went without any manual logging.<p>Traditional time trackers tell you "3 hours in Chrome" which is not very helpful. Dayflow actually understands if you're reading documentation, debugging code, or scrolling HN. Instead of "Chrome: 3 hours", you get "Reviewed PR comments: 45min", "Read HN thread about Rust: 20min", "Debugged auth flow: 1.5hr".<p>I was an early Rewind user but rarely used the retrieval feature. I built Dayflow because I saw other interesting uses for screen data. I find that it helps me stay on track while working - I check it every few hours and make sure I’m spending my time the way I intended - if I’m not, I try to course correct.<p>Here’s what you need to know about privacy:<p>- Run 100% locally using qwen2.5-vl-3b (~4GB model)<p>- No cloud uploads, no account<p>- Full source available under MIT license (<a href="https://github.com/JerryZLiu/Dayflow" rel="nofollow">https://github.com/JerryZLiu/Dayflow</a>)<p>- Optional: BYO Gemini API key for better quality (stored in Keychain, with free-tier workaround to prevent training on your data)<p>The tech stack is pretty simple, SwiftUI with a local sqlite DB. Uses native macOS apis for efficient screen captures. Since most people who run LLMs locally already have their tool of choice (Ollama, LLMStudio, etc.), I decided to not embed an LLM into Dayflow.<p>By far the biggest challenge was adapting from SOTA vision models like Gemini 2.5 Pro to small, local models. My constraints were that it had to take up <4GB of ram and have vision capabilities. I had to do a lot of evals to figure out that Qwen2.5VL-3B was the best balance of size and quality, but there was still a sizable tradeoff in quality that I had to accept. I also got creative with sampling rates and prompt chunking to deal with the 100x smaller context window. Processing a 15 minute segment takes ~32 local LLM calls vs 2 Gemini calls!<p>Here’s what I’m working on next:<p>Distillation: Using Gemini's high-quality outputs as training data to teach a local model the patterns it needs, hopefully closing the quality gap.<p>Custom dashboards where you can track answers to any question like "How long did I spend on HN?" or "Hours until my first deep work session of the day<p>I'd love to hear your thoughts, especially if you've struggled with productivity tracking or have ideas for what you'd want from a tool like this.

Show HN: Ggc – A Git CLI tool written in Go with interactive UI

A while ago I shared an early version of ggc, a Git helper I built in Go. Since then the project has grown quite a bit, and I’d love to share the latest updates (v6.0).<p>Repo: <a href="https://github.com/bmf-san/ggc" rel="nofollow">https://github.com/bmf-san/ggc</a><p>Install: - macOS/Linux: `brew install ggc` - Go: `go install github.com/bmf-san/ggc/v6@latest` - Homebrew: `brew install ggc` - Or grab binaries: <a href="https://github.com/bmf-san/ggc/releases" rel="nofollow">https://github.com/bmf-san/ggc/releases</a><p>Features: Dual modes: Traditional CLI commands (ggc add, etc.) and interactive mode (launch with just ggc) Intuitive command structure: Simplified interface for common Git operations Incremental search UI: Quickly find and execute commands with real-time filtering Fast and lightweight: Implemented in Go with minimal dependencies Shell completions: Included for Bash, Zsh, and Fish Custom aliases: Chain multiple commands with user-defined aliases Cross-platform: Works on macOS, Linux, and Windows<p>Technical details: Built with Go standard library and minimal external packages Supports 50+ Git operations (add, commit, branch, pull, etc.)<p>I'd appreciate any feedback or contributions!

Show HN: Ggc – A Git CLI tool written in Go with interactive UI

A while ago I shared an early version of ggc, a Git helper I built in Go. Since then the project has grown quite a bit, and I’d love to share the latest updates (v6.0).<p>Repo: <a href="https://github.com/bmf-san/ggc" rel="nofollow">https://github.com/bmf-san/ggc</a><p>Install: - macOS/Linux: `brew install ggc` - Go: `go install github.com/bmf-san/ggc/v6@latest` - Homebrew: `brew install ggc` - Or grab binaries: <a href="https://github.com/bmf-san/ggc/releases" rel="nofollow">https://github.com/bmf-san/ggc/releases</a><p>Features: Dual modes: Traditional CLI commands (ggc add, etc.) and interactive mode (launch with just ggc) Intuitive command structure: Simplified interface for common Git operations Incremental search UI: Quickly find and execute commands with real-time filtering Fast and lightweight: Implemented in Go with minimal dependencies Shell completions: Included for Bash, Zsh, and Fish Custom aliases: Chain multiple commands with user-defined aliases Cross-platform: Works on macOS, Linux, and Windows<p>Technical details: Built with Go standard library and minimal external packages Supports 50+ Git operations (add, commit, branch, pull, etc.)<p>I'd appreciate any feedback or contributions!

Show HN: Ggc – A Git CLI tool written in Go with interactive UI

A while ago I shared an early version of ggc, a Git helper I built in Go. Since then the project has grown quite a bit, and I’d love to share the latest updates (v6.0).<p>Repo: <a href="https://github.com/bmf-san/ggc" rel="nofollow">https://github.com/bmf-san/ggc</a><p>Install: - macOS/Linux: `brew install ggc` - Go: `go install github.com/bmf-san/ggc/v6@latest` - Homebrew: `brew install ggc` - Or grab binaries: <a href="https://github.com/bmf-san/ggc/releases" rel="nofollow">https://github.com/bmf-san/ggc/releases</a><p>Features: Dual modes: Traditional CLI commands (ggc add, etc.) and interactive mode (launch with just ggc) Intuitive command structure: Simplified interface for common Git operations Incremental search UI: Quickly find and execute commands with real-time filtering Fast and lightweight: Implemented in Go with minimal dependencies Shell completions: Included for Bash, Zsh, and Fish Custom aliases: Chain multiple commands with user-defined aliases Cross-platform: Works on macOS, Linux, and Windows<p>Technical details: Built with Go standard library and minimal external packages Supports 50+ Git operations (add, commit, branch, pull, etc.)<p>I'd appreciate any feedback or contributions!

Show HN: FlyCode – Recover Stripe payments by automatically using backup cards

We built FlyCode after seeing subscription businesses lose ~35% of recurring revenue each year to failed payments — even when customers had other valid cards on file.<p>*The problem:* When a customer's primary card fails, Stripe retries a few times then cancels the subscription. If that customer has a backup card, it isn’t tried. At least 20% of active customers have more than one card on file, which means a lot of preventable churn.<p>*Our solution:* FlyCode automatically identifies if a customer has other valid cards on file and retries them when a subscription payment fails. You can configure when these retries happen during the dunning period (beginning, middle, end) and define validity rules (e.g. “card was used in last 180 days”). It’s a Stripe app — no code changes needed.<p>We've seen 18%-20% higher recovery rates from our core retry engine, plus another 5–10% from using backup cards. Importantly, there was no increase in refunds or chargebacks — in fact, rates were lower than merchant averages. Big companies like Microsoft and Amazon already do this internally; we wanted to make the same capability accessible to smaller SaaS teams.<p>*Under the hood:* FlyCode monitors for failed invoices, checks for available backup methods via Stripe’s PaymentMethod API, and systematically retries in a way that avoids service disruption or manual workflows.<p>We’re Jake, Etai, and Tzachi — we previously built payment recovery systems at startups and enterprises, which is how we discovered this gap.<p>You can try it here: [<a href="https://www.flycode.com/stripe">https://www.flycode.com/stripe</a>]<p>We’d love feedback from anyone dealing with subscription payment failures. What’s been your experience with involuntary churn? Have you considered leveraging backup payment methods?

Show HN: FlyCode – Recover Stripe payments by automatically using backup cards

We built FlyCode after seeing subscription businesses lose ~35% of recurring revenue each year to failed payments — even when customers had other valid cards on file.<p>*The problem:* When a customer's primary card fails, Stripe retries a few times then cancels the subscription. If that customer has a backup card, it isn’t tried. At least 20% of active customers have more than one card on file, which means a lot of preventable churn.<p>*Our solution:* FlyCode automatically identifies if a customer has other valid cards on file and retries them when a subscription payment fails. You can configure when these retries happen during the dunning period (beginning, middle, end) and define validity rules (e.g. “card was used in last 180 days”). It’s a Stripe app — no code changes needed.<p>We've seen 18%-20% higher recovery rates from our core retry engine, plus another 5–10% from using backup cards. Importantly, there was no increase in refunds or chargebacks — in fact, rates were lower than merchant averages. Big companies like Microsoft and Amazon already do this internally; we wanted to make the same capability accessible to smaller SaaS teams.<p>*Under the hood:* FlyCode monitors for failed invoices, checks for available backup methods via Stripe’s PaymentMethod API, and systematically retries in a way that avoids service disruption or manual workflows.<p>We’re Jake, Etai, and Tzachi — we previously built payment recovery systems at startups and enterprises, which is how we discovered this gap.<p>You can try it here: [<a href="https://www.flycode.com/stripe">https://www.flycode.com/stripe</a>]<p>We’d love feedback from anyone dealing with subscription payment failures. What’s been your experience with involuntary churn? Have you considered leveraging backup payment methods?

Show HN: The Blots Programming Language

I've been working on this small, slightly weird expression-oriented programming language for a little while now and feel ready to share it with others. I use it pretty often now in my day-to-day and work life, as a scratchpad for doing a bit of quick math or picking some pieces of data out of a JSON payload.<p>Would really appreciate any feedback about the syntax, docs, features that are glaringly missing, etc. Before anybody mentions it: I know the performance is pretty lousy when dealing with a lot of data. If you can believe it, the runtime is about 100x faster than it used to be! Long term I'd like to switch to a proper bytecode interpreter, but so far performance has been Good Enough for my use cases.<p>Thanks for taking a look!