The best Hacker News stories from Show from the past day

Go back

Latest posts:

Show HN: Browser AI agent platform designed for reliability

We’re very excited to share something we’ve been building. Notte <a href="https://www.notte.cc/">https://www.notte.cc/</a> is a full-stack browser agent platform built to reliably automate a wide range of workflows.<p>Browser agents aren’t new, but what is still hard is covering real-world flows reliably. The inspiration for Notte was to make a full-featured platform that bridges the agent reliability gap. We’ve packaged everything via a singe API for ease of use:<p>- Site Interactions - Observe website states, scrape data and execute actions<p>- Structured Output - Get data in your exact format with Pydantic models<p>- Stealth browser sessions - built-in CAPTCHA solving, proxies, and anti-detection<p>- Hybrid workflows - Combine scripting and AI agents to reduce costs and improve reliability<p>- Secrets vaults - Credential management to store emails, passwords, MFA tokens, SSO, etc.<p>- Digital personas - Digital identities with unique emails, phones for account creation workflows<p>With these tools, Notte allows you to automate difficult tasks like account creation, form filling, work on authenticated dashboards. Close compatibility with Playwright allows you to cut LLM costs and improve execution speed by mixing web automation primitives and include agents only for specific parts that require reasoning and adaptability.<p>Here’s a short YouTube demo: <a href="https://www.youtube.com/watch?v=b1CzmfpdzaQ" rel="nofollow">https://www.youtube.com/watch?v=b1CzmfpdzaQ</a><p>If any of this sounds interesting, you can run your first agent following our quickstart on GitHub <a href="https://github.com/nottelabs/notte" rel="nofollow">https://github.com/nottelabs/notte</a>. Or play around with our free plan through our Notte Console: <a href="https://console.notte.cc/">https://console.notte.cc/</a><p>We’d love to hear if there’s anything else required before you’d try or trust it on your own workflows :)

Show HN: Browser AI agent platform designed for reliability

We’re very excited to share something we’ve been building. Notte <a href="https://www.notte.cc/">https://www.notte.cc/</a> is a full-stack browser agent platform built to reliably automate a wide range of workflows.<p>Browser agents aren’t new, but what is still hard is covering real-world flows reliably. The inspiration for Notte was to make a full-featured platform that bridges the agent reliability gap. We’ve packaged everything via a singe API for ease of use:<p>- Site Interactions - Observe website states, scrape data and execute actions<p>- Structured Output - Get data in your exact format with Pydantic models<p>- Stealth browser sessions - built-in CAPTCHA solving, proxies, and anti-detection<p>- Hybrid workflows - Combine scripting and AI agents to reduce costs and improve reliability<p>- Secrets vaults - Credential management to store emails, passwords, MFA tokens, SSO, etc.<p>- Digital personas - Digital identities with unique emails, phones for account creation workflows<p>With these tools, Notte allows you to automate difficult tasks like account creation, form filling, work on authenticated dashboards. Close compatibility with Playwright allows you to cut LLM costs and improve execution speed by mixing web automation primitives and include agents only for specific parts that require reasoning and adaptability.<p>Here’s a short YouTube demo: <a href="https://www.youtube.com/watch?v=b1CzmfpdzaQ" rel="nofollow">https://www.youtube.com/watch?v=b1CzmfpdzaQ</a><p>If any of this sounds interesting, you can run your first agent following our quickstart on GitHub <a href="https://github.com/nottelabs/notte" rel="nofollow">https://github.com/nottelabs/notte</a>. Or play around with our free plan through our Notte Console: <a href="https://console.notte.cc/">https://console.notte.cc/</a><p>We’d love to hear if there’s anything else required before you’d try or trust it on your own workflows :)

Show HN: Browser AI agent platform designed for reliability

We’re very excited to share something we’ve been building. Notte <a href="https://www.notte.cc/">https://www.notte.cc/</a> is a full-stack browser agent platform built to reliably automate a wide range of workflows.<p>Browser agents aren’t new, but what is still hard is covering real-world flows reliably. The inspiration for Notte was to make a full-featured platform that bridges the agent reliability gap. We’ve packaged everything via a singe API for ease of use:<p>- Site Interactions - Observe website states, scrape data and execute actions<p>- Structured Output - Get data in your exact format with Pydantic models<p>- Stealth browser sessions - built-in CAPTCHA solving, proxies, and anti-detection<p>- Hybrid workflows - Combine scripting and AI agents to reduce costs and improve reliability<p>- Secrets vaults - Credential management to store emails, passwords, MFA tokens, SSO, etc.<p>- Digital personas - Digital identities with unique emails, phones for account creation workflows<p>With these tools, Notte allows you to automate difficult tasks like account creation, form filling, work on authenticated dashboards. Close compatibility with Playwright allows you to cut LLM costs and improve execution speed by mixing web automation primitives and include agents only for specific parts that require reasoning and adaptability.<p>Here’s a short YouTube demo: <a href="https://www.youtube.com/watch?v=b1CzmfpdzaQ" rel="nofollow">https://www.youtube.com/watch?v=b1CzmfpdzaQ</a><p>If any of this sounds interesting, you can run your first agent following our quickstart on GitHub <a href="https://github.com/nottelabs/notte" rel="nofollow">https://github.com/nottelabs/notte</a>. Or play around with our free plan through our Notte Console: <a href="https://console.notte.cc/">https://console.notte.cc/</a><p>We’d love to hear if there’s anything else required before you’d try or trust it on your own workflows :)

Show HN: Browser AI agent platform designed for reliability

We’re very excited to share something we’ve been building. Notte <a href="https://www.notte.cc/">https://www.notte.cc/</a> is a full-stack browser agent platform built to reliably automate a wide range of workflows.<p>Browser agents aren’t new, but what is still hard is covering real-world flows reliably. The inspiration for Notte was to make a full-featured platform that bridges the agent reliability gap. We’ve packaged everything via a singe API for ease of use:<p>- Site Interactions - Observe website states, scrape data and execute actions<p>- Structured Output - Get data in your exact format with Pydantic models<p>- Stealth browser sessions - built-in CAPTCHA solving, proxies, and anti-detection<p>- Hybrid workflows - Combine scripting and AI agents to reduce costs and improve reliability<p>- Secrets vaults - Credential management to store emails, passwords, MFA tokens, SSO, etc.<p>- Digital personas - Digital identities with unique emails, phones for account creation workflows<p>With these tools, Notte allows you to automate difficult tasks like account creation, form filling, work on authenticated dashboards. Close compatibility with Playwright allows you to cut LLM costs and improve execution speed by mixing web automation primitives and include agents only for specific parts that require reasoning and adaptability.<p>Here’s a short YouTube demo: <a href="https://www.youtube.com/watch?v=b1CzmfpdzaQ" rel="nofollow">https://www.youtube.com/watch?v=b1CzmfpdzaQ</a><p>If any of this sounds interesting, you can run your first agent following our quickstart on GitHub <a href="https://github.com/nottelabs/notte" rel="nofollow">https://github.com/nottelabs/notte</a>. Or play around with our free plan through our Notte Console: <a href="https://console.notte.cc/">https://console.notte.cc/</a><p>We’d love to hear if there’s anything else required before you’d try or trust it on your own workflows :)

Show HN: Browser AI agent platform designed for reliability

We’re very excited to share something we’ve been building. Notte <a href="https://www.notte.cc/">https://www.notte.cc/</a> is a full-stack browser agent platform built to reliably automate a wide range of workflows.<p>Browser agents aren’t new, but what is still hard is covering real-world flows reliably. The inspiration for Notte was to make a full-featured platform that bridges the agent reliability gap. We’ve packaged everything via a singe API for ease of use:<p>- Site Interactions - Observe website states, scrape data and execute actions<p>- Structured Output - Get data in your exact format with Pydantic models<p>- Stealth browser sessions - built-in CAPTCHA solving, proxies, and anti-detection<p>- Hybrid workflows - Combine scripting and AI agents to reduce costs and improve reliability<p>- Secrets vaults - Credential management to store emails, passwords, MFA tokens, SSO, etc.<p>- Digital personas - Digital identities with unique emails, phones for account creation workflows<p>With these tools, Notte allows you to automate difficult tasks like account creation, form filling, work on authenticated dashboards. Close compatibility with Playwright allows you to cut LLM costs and improve execution speed by mixing web automation primitives and include agents only for specific parts that require reasoning and adaptability.<p>Here’s a short YouTube demo: <a href="https://www.youtube.com/watch?v=b1CzmfpdzaQ" rel="nofollow">https://www.youtube.com/watch?v=b1CzmfpdzaQ</a><p>If any of this sounds interesting, you can run your first agent following our quickstart on GitHub <a href="https://github.com/nottelabs/notte" rel="nofollow">https://github.com/nottelabs/notte</a>. Or play around with our free plan through our Notte Console: <a href="https://console.notte.cc/">https://console.notte.cc/</a><p>We’d love to hear if there’s anything else required before you’d try or trust it on your own workflows :)

Show HN: Stasher – Burn-after-read secrets from the CLI, no server, no trust

Stasher is a tiny CLI tool that lets you share encrypted secrets that burn after reading — no accounts, no logins, no servers to trust.<p>I built it because I just wanted to share a password. Not spin up infra. Not register for some "secure" web app. Not trust Slack threads. Just send a secret.<p>Secrets are encrypted client-side with AES-256-GCM. You get a `uuid:key` token to share. Once someone reads it, it's gone. If they don't read it in 10 minutes, it expires and deleted.<p>Everything is verifiable. Every release is signed, SLSA-attested, SBOM-included, and logged in the Rekor transparency log. Every line of code is public.<p>There's also a browser-based companion: <a href="https://app.stasher.dev" rel="nofollow">https://app.stasher.dev</a> — works in a sandboxed popup using the same encrypted model. Share from the terminal, pick up in the browser.<p>No data stored unencrypted. No metadata. No logs. No surveillance.<p>---<p>GitHub (CLI): <a href="https://github.com/stasher-dev/stasher-cli" rel="nofollow">https://github.com/stasher-dev/stasher-cli</a> GitHub (App): <a href="https://github.com/stasher-dev/stasher-app" rel="nofollow">https://github.com/stasher-dev/stasher-app</a> API (Cloudflare Worker): <a href="https://github.com/stasher-dev/stasher-api" rel="nofollow">https://github.com/stasher-dev/stasher-api</a> CI/CD (Open): <a href="https://github.com/stasher-dev/stasher-ci" rel="nofollow">https://github.com/stasher-dev/stasher-ci</a> NPM: <a href="https://www.npmjs.com/package/stasher-cli" rel="nofollow">https://www.npmjs.com/package/stasher-cli</a> Website: <a href="https://stasher.dev" rel="nofollow">https://stasher.dev</a> Browser App: <a href="https://app.stasher.dev" rel="nofollow">https://app.stasher.dev</a> (runs in sandbox from <a href="https://dev.stasher" rel="nofollow">https://dev.stasher</a>)<p>Built with Cloudflare Workers, KV, and Durable Objects. All code open, auditable, and signed.<p>Try it:<p>```bash npx enstash "vault code is 1234#" npx destash "uuid:base64key"<p>thanks for reading

Show HN: Stasher – Burn-after-read secrets from the CLI, no server, no trust

Stasher is a tiny CLI tool that lets you share encrypted secrets that burn after reading — no accounts, no logins, no servers to trust.<p>I built it because I just wanted to share a password. Not spin up infra. Not register for some "secure" web app. Not trust Slack threads. Just send a secret.<p>Secrets are encrypted client-side with AES-256-GCM. You get a `uuid:key` token to share. Once someone reads it, it's gone. If they don't read it in 10 minutes, it expires and deleted.<p>Everything is verifiable. Every release is signed, SLSA-attested, SBOM-included, and logged in the Rekor transparency log. Every line of code is public.<p>There's also a browser-based companion: <a href="https://app.stasher.dev" rel="nofollow">https://app.stasher.dev</a> — works in a sandboxed popup using the same encrypted model. Share from the terminal, pick up in the browser.<p>No data stored unencrypted. No metadata. No logs. No surveillance.<p>---<p>GitHub (CLI): <a href="https://github.com/stasher-dev/stasher-cli" rel="nofollow">https://github.com/stasher-dev/stasher-cli</a> GitHub (App): <a href="https://github.com/stasher-dev/stasher-app" rel="nofollow">https://github.com/stasher-dev/stasher-app</a> API (Cloudflare Worker): <a href="https://github.com/stasher-dev/stasher-api" rel="nofollow">https://github.com/stasher-dev/stasher-api</a> CI/CD (Open): <a href="https://github.com/stasher-dev/stasher-ci" rel="nofollow">https://github.com/stasher-dev/stasher-ci</a> NPM: <a href="https://www.npmjs.com/package/stasher-cli" rel="nofollow">https://www.npmjs.com/package/stasher-cli</a> Website: <a href="https://stasher.dev" rel="nofollow">https://stasher.dev</a> Browser App: <a href="https://app.stasher.dev" rel="nofollow">https://app.stasher.dev</a> (runs in sandbox from <a href="https://dev.stasher" rel="nofollow">https://dev.stasher</a>)<p>Built with Cloudflare Workers, KV, and Durable Objects. All code open, auditable, and signed.<p>Try it:<p>```bash npx enstash "vault code is 1234#" npx destash "uuid:base64key"<p>thanks for reading

Show HN: Stasher – Burn-after-read secrets from the CLI, no server, no trust

Stasher is a tiny CLI tool that lets you share encrypted secrets that burn after reading — no accounts, no logins, no servers to trust.<p>I built it because I just wanted to share a password. Not spin up infra. Not register for some "secure" web app. Not trust Slack threads. Just send a secret.<p>Secrets are encrypted client-side with AES-256-GCM. You get a `uuid:key` token to share. Once someone reads it, it's gone. If they don't read it in 10 minutes, it expires and deleted.<p>Everything is verifiable. Every release is signed, SLSA-attested, SBOM-included, and logged in the Rekor transparency log. Every line of code is public.<p>There's also a browser-based companion: <a href="https://app.stasher.dev" rel="nofollow">https://app.stasher.dev</a> — works in a sandboxed popup using the same encrypted model. Share from the terminal, pick up in the browser.<p>No data stored unencrypted. No metadata. No logs. No surveillance.<p>---<p>GitHub (CLI): <a href="https://github.com/stasher-dev/stasher-cli" rel="nofollow">https://github.com/stasher-dev/stasher-cli</a> GitHub (App): <a href="https://github.com/stasher-dev/stasher-app" rel="nofollow">https://github.com/stasher-dev/stasher-app</a> API (Cloudflare Worker): <a href="https://github.com/stasher-dev/stasher-api" rel="nofollow">https://github.com/stasher-dev/stasher-api</a> CI/CD (Open): <a href="https://github.com/stasher-dev/stasher-ci" rel="nofollow">https://github.com/stasher-dev/stasher-ci</a> NPM: <a href="https://www.npmjs.com/package/stasher-cli" rel="nofollow">https://www.npmjs.com/package/stasher-cli</a> Website: <a href="https://stasher.dev" rel="nofollow">https://stasher.dev</a> Browser App: <a href="https://app.stasher.dev" rel="nofollow">https://app.stasher.dev</a> (runs in sandbox from <a href="https://dev.stasher" rel="nofollow">https://dev.stasher</a>)<p>Built with Cloudflare Workers, KV, and Durable Objects. All code open, auditable, and signed.<p>Try it:<p>```bash npx enstash "vault code is 1234#" npx destash "uuid:base64key"<p>thanks for reading

Show HN: Stasher – Burn-after-read secrets from the CLI, no server, no trust

Stasher is a tiny CLI tool that lets you share encrypted secrets that burn after reading — no accounts, no logins, no servers to trust.<p>I built it because I just wanted to share a password. Not spin up infra. Not register for some "secure" web app. Not trust Slack threads. Just send a secret.<p>Secrets are encrypted client-side with AES-256-GCM. You get a `uuid:key` token to share. Once someone reads it, it's gone. If they don't read it in 10 minutes, it expires and deleted.<p>Everything is verifiable. Every release is signed, SLSA-attested, SBOM-included, and logged in the Rekor transparency log. Every line of code is public.<p>There's also a browser-based companion: <a href="https://app.stasher.dev" rel="nofollow">https://app.stasher.dev</a> — works in a sandboxed popup using the same encrypted model. Share from the terminal, pick up in the browser.<p>No data stored unencrypted. No metadata. No logs. No surveillance.<p>---<p>GitHub (CLI): <a href="https://github.com/stasher-dev/stasher-cli" rel="nofollow">https://github.com/stasher-dev/stasher-cli</a> GitHub (App): <a href="https://github.com/stasher-dev/stasher-app" rel="nofollow">https://github.com/stasher-dev/stasher-app</a> API (Cloudflare Worker): <a href="https://github.com/stasher-dev/stasher-api" rel="nofollow">https://github.com/stasher-dev/stasher-api</a> CI/CD (Open): <a href="https://github.com/stasher-dev/stasher-ci" rel="nofollow">https://github.com/stasher-dev/stasher-ci</a> NPM: <a href="https://www.npmjs.com/package/stasher-cli" rel="nofollow">https://www.npmjs.com/package/stasher-cli</a> Website: <a href="https://stasher.dev" rel="nofollow">https://stasher.dev</a> Browser App: <a href="https://app.stasher.dev" rel="nofollow">https://app.stasher.dev</a> (runs in sandbox from <a href="https://dev.stasher" rel="nofollow">https://dev.stasher</a>)<p>Built with Cloudflare Workers, KV, and Durable Objects. All code open, auditable, and signed.<p>Try it:<p>```bash npx enstash "vault code is 1234#" npx destash "uuid:base64key"<p>thanks for reading

Show HN: Octofriend, a cute coding agent that can swap between GPT-5 and Claude

Hey HN! We're shipping Octofriend today, a cute coding assistant that can swap between GPT-5, Claude, local or open-source LLMs, etc mid-conversation as needed. It handles reasoning tokens (including encrypted ones from OpenAI and Anthropic) really well, and includes a couple of custom-trained ML models to fix minor diff edit and JSON encoding errors that we've also open-sourced. Have fun!

Show HN: Octofriend, a cute coding agent that can swap between GPT-5 and Claude

Hey HN! We're shipping Octofriend today, a cute coding assistant that can swap between GPT-5, Claude, local or open-source LLMs, etc mid-conversation as needed. It handles reasoning tokens (including encrypted ones from OpenAI and Anthropic) really well, and includes a couple of custom-trained ML models to fix minor diff edit and JSON encoding errors that we've also open-sourced. Have fun!

Show HN: Octofriend, a cute coding agent that can swap between GPT-5 and Claude

Hey HN! We're shipping Octofriend today, a cute coding assistant that can swap between GPT-5, Claude, local or open-source LLMs, etc mid-conversation as needed. It handles reasoning tokens (including encrypted ones from OpenAI and Anthropic) really well, and includes a couple of custom-trained ML models to fix minor diff edit and JSON encoding errors that we've also open-sourced. Have fun!

Show HN: Octofriend, a cute coding agent that can swap between GPT-5 and Claude

Hey HN! We're shipping Octofriend today, a cute coding assistant that can swap between GPT-5, Claude, local or open-source LLMs, etc mid-conversation as needed. It handles reasoning tokens (including encrypted ones from OpenAI and Anthropic) really well, and includes a couple of custom-trained ML models to fix minor diff edit and JSON encoding errors that we've also open-sourced. Have fun!

Show HN: Sinkzone DNS – Forwarder that blocks everything except your allowlist

Most site blockers work by blacklisting distractions. That never worked for me, the internet is too big, and there’s always something new to waste time on.<p>I wanted the opposite: allowlist‑only browsing. Block everything by default, and explicitly allow only what I need.<p>So I built Sinkzone: a local DNS forwarder with two modes:<p>Monitor mode: lets all traffic through, but logs every domain so you can decide what to allow.<p>Focus mode: only allowlisted domains resolve; everything else is blocked (NXDOMAIN).<p>It’s open source, written in Go, and runs locally on macOS, Linux, and Windows. Works a bit like Pi‑hole, but instead of blocking ads, it blocks everything unless you say otherwise.<p>I’m curious if this would be useful in your workflow. If you try it, please let me know what breaks, what works well, and what you’d improve.

Show HN: Sinkzone DNS – Forwarder that blocks everything except your allowlist

Most site blockers work by blacklisting distractions. That never worked for me, the internet is too big, and there’s always something new to waste time on.<p>I wanted the opposite: allowlist‑only browsing. Block everything by default, and explicitly allow only what I need.<p>So I built Sinkzone: a local DNS forwarder with two modes:<p>Monitor mode: lets all traffic through, but logs every domain so you can decide what to allow.<p>Focus mode: only allowlisted domains resolve; everything else is blocked (NXDOMAIN).<p>It’s open source, written in Go, and runs locally on macOS, Linux, and Windows. Works a bit like Pi‑hole, but instead of blocking ads, it blocks everything unless you say otherwise.<p>I’m curious if this would be useful in your workflow. If you try it, please let me know what breaks, what works well, and what you’d improve.

Show HN: Sinkzone DNS – Forwarder that blocks everything except your allowlist

Most site blockers work by blacklisting distractions. That never worked for me, the internet is too big, and there’s always something new to waste time on.<p>I wanted the opposite: allowlist‑only browsing. Block everything by default, and explicitly allow only what I need.<p>So I built Sinkzone: a local DNS forwarder with two modes:<p>Monitor mode: lets all traffic through, but logs every domain so you can decide what to allow.<p>Focus mode: only allowlisted domains resolve; everything else is blocked (NXDOMAIN).<p>It’s open source, written in Go, and runs locally on macOS, Linux, and Windows. Works a bit like Pi‑hole, but instead of blocking ads, it blocks everything unless you say otherwise.<p>I’m curious if this would be useful in your workflow. If you try it, please let me know what breaks, what works well, and what you’d improve.

Show HN: Sinkzone DNS – Forwarder that blocks everything except your allowlist

Most site blockers work by blacklisting distractions. That never worked for me, the internet is too big, and there’s always something new to waste time on.<p>I wanted the opposite: allowlist‑only browsing. Block everything by default, and explicitly allow only what I need.<p>So I built Sinkzone: a local DNS forwarder with two modes:<p>Monitor mode: lets all traffic through, but logs every domain so you can decide what to allow.<p>Focus mode: only allowlisted domains resolve; everything else is blocked (NXDOMAIN).<p>It’s open source, written in Go, and runs locally on macOS, Linux, and Windows. Works a bit like Pi‑hole, but instead of blocking ads, it blocks everything unless you say otherwise.<p>I’m curious if this would be useful in your workflow. If you try it, please let me know what breaks, what works well, and what you’d improve.

Show HN: An open-source e-book reader for conversational reading with an LLM

Hi HN! I've been working on BookWith, an open-source e-book reader that integrates AI as your reading companion.<p>The problem: Traditional e-readers are passive. When you encounter something unclear, you have to context-switch to search for it. Your highlights and notes remain isolated, and you can't easily connect ideas across different books.<p>My solution: BookWith embeds an AI that maintains full context of what you're reading. It features:<p>- Context-aware AI chat: Ask questions about the current page/chapter and get instant answers<p>- AI podcast generation: Automatically converts book content into conversational podcasts using Google Cloud TTS<p>- Multi-layer memory system: Short-term (last 5 conversations), mid-term (summarized every 20), and long-term (vector search) memory that maintains continuity across reading sessions<p>- Smart annotations: 5-color highlighting system that AI can reference and analyze<p>Technical stack: Built as a fork of Flow (epub reader), with added LLM integration and vector database for semantic search. Supports multiple LLMs and languages (EN/JA/ZH).

Show HN: An open-source e-book reader for conversational reading with an LLM

Hi HN! I've been working on BookWith, an open-source e-book reader that integrates AI as your reading companion.<p>The problem: Traditional e-readers are passive. When you encounter something unclear, you have to context-switch to search for it. Your highlights and notes remain isolated, and you can't easily connect ideas across different books.<p>My solution: BookWith embeds an AI that maintains full context of what you're reading. It features:<p>- Context-aware AI chat: Ask questions about the current page/chapter and get instant answers<p>- AI podcast generation: Automatically converts book content into conversational podcasts using Google Cloud TTS<p>- Multi-layer memory system: Short-term (last 5 conversations), mid-term (summarized every 20), and long-term (vector search) memory that maintains continuity across reading sessions<p>- Smart annotations: 5-color highlighting system that AI can reference and analyze<p>Technical stack: Built as a fork of Flow (epub reader), with added LLM integration and vector database for semantic search. Supports multiple LLMs and languages (EN/JA/ZH).

Show HN: An open-source e-book reader for conversational reading with an LLM

Hi HN! I've been working on BookWith, an open-source e-book reader that integrates AI as your reading companion.<p>The problem: Traditional e-readers are passive. When you encounter something unclear, you have to context-switch to search for it. Your highlights and notes remain isolated, and you can't easily connect ideas across different books.<p>My solution: BookWith embeds an AI that maintains full context of what you're reading. It features:<p>- Context-aware AI chat: Ask questions about the current page/chapter and get instant answers<p>- AI podcast generation: Automatically converts book content into conversational podcasts using Google Cloud TTS<p>- Multi-layer memory system: Short-term (last 5 conversations), mid-term (summarized every 20), and long-term (vector search) memory that maintains continuity across reading sessions<p>- Smart annotations: 5-color highlighting system that AI can reference and analyze<p>Technical stack: Built as a fork of Flow (epub reader), with added LLM integration and vector database for semantic search. Supports multiple LLMs and languages (EN/JA/ZH).

< 1 2 3 ... 18 19 20 21 22 ... 866 867 868 >