frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

RageDetector – detects aggressive typing and forces me to calm down

https://github.com/AI-Architechs/RageDetector
2•karan_dev•3m ago•1 comments

Lentando Private Habit Tracker

https://frankforce.com/lentando-%f0%9f%90%a2-private-habit-and-substance-tracker/
1•memalign•7m ago•0 comments

Why Europe doesn't have a Tesla

https://worksinprogress.co/issue/why-europe-doesnt-have-a-tesla/
1•trojanalert•8m ago•1 comments

SnkvDB – Single-header ACID KV store using SQLite's B-Tree engine

https://github.com/hash-anu/snkv
1•usefulcat•8m ago•0 comments

Software Engineering in 2026

https://twitter.com/Adityapandeydev/status/2023620303126229276
1•keepamovin•8m ago•1 comments

Ask HN: Is the definition of AI that it can fool people?

2•WhatsTheBigIdea•10m ago•0 comments

Choose Your Fictions Well (2010)

http://henryjenkins.org/blog/2010/04/choose_your_ficitons_well.html
1•1970-01-01•10m ago•0 comments

Tell HN: Google Allegedly Sent NSFW "Grok" Notification to People

1•surprisetalk•11m ago•0 comments

If AI Agents Do the Work, Who Pays for the Seat?

https://www.linkedin.com/pulse/ai-agents-do-work-who-pays-seat-c-max-magee-eygwe
2•wawayanda•22m ago•0 comments

Agent Skills Hub – Security first directory for AI agent skills and MCP

https://agentskillshub.dev/
1•cana2026•23m ago•0 comments

Show HN: TerminalRant – Mastodon for developers who live in the terminal

https://github.com/CrestNiraj12/TerminalRant
1•crestniraj•28m ago•0 comments

Show HN: QuickStaging – AI virtual staging tool built by a 19yo student

https://getquickstaging.com
1•kadiran•30m ago•1 comments

From Claude Code to Figma

https://www.figma.com/blog/introducing-claude-code-to-figma/
1•msolujic•35m ago•0 comments

Wi-Fi 7's Best Feature Doesn't Work (Yet) [video]

https://www.youtube.com/watch?v=-5o_Qu3XToQ
1•pss314•37m ago•0 comments

Show HN: StatusPing – Uptime monitoring for $9/mo

https://statusping.dev
1•pdtech_apps•39m ago•1 comments

Pidgin Plugins

https://pidgin.im/plugins/
1•xnx•40m ago•0 comments

How the Olympics Are Mixed Live 4k Miles Away [video]

https://www.youtube.com/watch?v=ZAJSaN8Z3Ik
1•geerlingguy•48m ago•0 comments

Meshcore IRC Bridge

https://github.com/daniel-j-h/meshcore-irc-bridge
1•rickcarlino•48m ago•0 comments

Copper-rs the deterministic OS for robotics gets full observability

https://www.copper-robotics.com/whats-new/unlock-full-observability-with-copper-rs-v013
2•gbin•50m ago•0 comments

Show HN: Verified 16.7M Mac chip architecture on $60 Android phone

2•jltackett•52m ago•0 comments

Multi-Language MCP Server Performance Benchmark

https://www.tmdevlab.com/mcp-server-performance-benchmark.html
1•oshanz•55m ago•1 comments

Stop building generic AI chatbots: 45% of support leaders are ahead

https://ideatolaunch.co/blog/why-the-world-doesn-t-need-another-generic-ai-chatbot-and-how-to-bui...
2•DonAj•58m ago•0 comments

A Local-Algebraic Route to Emergent Gravity (100 Pages)

1•berndtzl•1h ago•1 comments

Cultivating Praxia

https://tasshin.com/blog/cultivating-praxia/
2•tasshin•1h ago•0 comments

Managing Docker Composes via GitOps

1•anuragxd•1h ago•0 comments

An update on upki

https://discourse.ubuntu.com/t/an-update-on-upki/77063
3•pabs3•1h ago•0 comments

Google trying to recover footage from other Guthrie home cameras

https://www.youtube.com/watch?v=658FsUNHZ0Q
1•busymom0•1h ago•2 comments

Way to Understand the Irish Economy

https://stephenkinsella.substack.com/p/the-best-way-to-understand-the-irish
2•paulpauper•1h ago•0 comments

Mature Cultural Desire

https://www.overcomingbias.com/p/mature-cultural-desire
2•paulpauper•1h ago•0 comments

Technology has changed the world in my lifetime

https://www.noahpinion.blog/p/how-technology-has-already-changed
1•paulpauper•1h ago•0 comments
Open in hackernews

Show HN: AI pentester – verified exploits, $999/assessment

2•gauravbsinghal•1h ago
I spent 20 years in security, most recently leading 100+ engineers at AWS building pentesting infrastructure across thousands of services. The same problem everywhere: pentests take weeks, cost $15-50k, and the results are stale before they ship.

I built Cipher to fix that. It's an AI agent that reasons like an attacker — maps the target, finds vulnerabilities, chains them into exploits, and proves they're real. Every finding ships with a reproducible Python script. If the script doesn't break your system, we don't report it.

How it works: Cipher defines security invariants ("User A can't access User B's data"), then multiple agents attack in parallel to violate them. A separate judge agent tries to disprove every finding — if it can't reproduce the exploit 3 times, the finding dies. You never see it.

$999 per assessment. Results in ~2 hours. Unlimited retesting.

Honest limitations: complex multi-step auth flows (SSO with MFA) still need manual setup like providing JWT credentials. We're working on it.

I'll run Cipher free for the first 15 HN readers who want to try it. Drop your email or sign up at https://apxlabs.ai/. Happy to answer any questions about the approach.

Comments

tonetegeatinst•1h ago
Are you able to share what models or fine tuning you did for the agents?

I'm currently studying security in college, and most of my time is spent working on a good system card and premade prompts for certain situations like using nmap or burpsuite.

gauravbsinghal•1h ago
Great question. We use frontier models (Claude, Gemini class) without fine-tuning. The insight that changed everything for us: prompt engineering alone hits a ceiling fast for offensive security.

What matters more than the model:

1. Architecture over prompts. Cipher isn't one agent with a great prompt — it's multiple agents with distinct roles (recon, attack, verification) that coordinate. The "judge" agent that tries to disprove findings is more important than the attacker agent. 2. Tool use over reasoning. The model doesn't "know" how to pentest — it reasons about what tool to use next based on what it's learned so far. We give it real tools (not simulated ones) and let it chain them. 3. Invariant-based testing over checklist-based. Instead of "try SQLi on every input," Cipher defines security properties ("User A can't access User B's data") and tries to violate them. This catches logic bugs that no scanner finds.

Since you're studying security — the best thing you can do is get really good at manual pentesting first. Understanding why an attack chain works is what lets you build agents that reason about it. The prompts matter less than the mental model you encode into the system's architecture.

Happy to chat more — feel free to DM or join our Discord.