frontpage.
newsnewestaskshowjobs

Made with ♥ by @iamnishanth

Open Source @Github

fp.

Are AI agents ready for the workplace? A new benchmark raises doubts

https://techcrunch.com/2026/01/22/are-ai-agents-ready-for-the-workplace-a-new-benchmark-raises-do...
1•PaulHoule•2m ago•0 comments

Show HN: AI Watermark and Stego Scanner

https://ulrischa.github.io/AIWatermarkDetector/
1•ulrischa•2m ago•0 comments

Clarity vs. complexity: the invisible work of subtraction

https://www.alexscamp.com/p/clarity-vs-complexity-the-invisible
1•dovhyi•3m ago•0 comments

Solid-State Freezer Needs No Refrigerants

https://spectrum.ieee.org/subzero-elastocaloric-cooling
1•Brajeshwar•4m ago•0 comments

Ask HN: Will LLMs/AI Decrease Human Intelligence and Make Expertise a Commodity?

1•mc-0•5m ago•1 comments

From Zero to Hero: A Brief Introduction to Spring Boot

https://jcob-sikorski.github.io/me/writing/from-zero-to-hello-world-spring-boot
1•jcob_sikorski•5m ago•0 comments

NSA detected phone call between foreign intelligence and person close to Trump

https://www.theguardian.com/us-news/2026/feb/07/nsa-foreign-intelligence-trump-whistleblower
4•c420•6m ago•0 comments

How to Fake a Robotics Result

https://itcanthink.substack.com/p/how-to-fake-a-robotics-result
1•ai_critic•6m ago•0 comments

It's time for the world to boycott the US

https://www.aljazeera.com/opinions/2026/2/5/its-time-for-the-world-to-boycott-the-us
1•HotGarbage•7m ago•0 comments

Show HN: Semantic Search for terminal commands in the Browser (No Back end)

https://jslambda.github.io/tldr-vsearch/
1•jslambda•7m ago•1 comments

The AI CEO Experiment

https://yukicapital.com/blog/the-ai-ceo-experiment/
2•romainsimon•8m ago•0 comments

Speed up responses with fast mode

https://code.claude.com/docs/en/fast-mode
3•surprisetalk•12m ago•0 comments

MS-DOS game copy protection and cracks

https://www.dosdays.co.uk/topics/game_cracks.php
3•TheCraiggers•13m ago•0 comments

Updates on GNU/Hurd progress [video]

https://fosdem.org/2026/schedule/event/7FZXHF-updates_on_gnuhurd_progress_rump_drivers_64bit_smp_...
2•birdculture•14m ago•0 comments

Epstein took a photo of his 2015 dinner with Zuckerberg and Musk

https://xcancel.com/search?f=tweets&q=davenewworld_2%2Fstatus%2F2020128223850316274
7•doener•14m ago•2 comments

MyFlames: Visualize MySQL query execution plans as interactive FlameGraphs

https://github.com/vgrippa/myflames
1•tanelpoder•15m ago•0 comments

Show HN: LLM of Babel

https://clairefro.github.io/llm-of-babel/
1•marjipan200•15m ago•0 comments

A modern iperf3 alternative with a live TUI, multi-client server, QUIC support

https://github.com/lance0/xfr
3•tanelpoder•17m ago•0 comments

Famfamfam Silk icons – also with CSS spritesheet

https://github.com/legacy-icons/famfamfam-silk
1•thunderbong•17m ago•0 comments

Apple is the only Big Tech company whose capex declined last quarter

https://sherwood.news/tech/apple-is-the-only-big-tech-company-whose-capex-declined-last-quarter/
2•elsewhen•20m ago•0 comments

Reverse-Engineering Raiders of the Lost Ark for the Atari 2600

https://github.com/joshuanwalker/Raiders2600
2•todsacerdoti•22m ago•0 comments

Show HN: Deterministic NDJSON audit logs – v1.2 update (structural gaps)

https://github.com/yupme-bot/kernel-ndjson-proofs
1•Slaine•25m ago•0 comments

The Greater Copenhagen Region could be your friend's next career move

https://www.greatercphregion.com/friend-recruiter-program
2•mooreds•26m ago•0 comments

Do Not Confirm – Fiction by OpenClaw

https://thedailymolt.substack.com/p/do-not-confirm
1•jamesjyu•26m ago•0 comments

The Analytical Profile of Peas

https://www.fossanalytics.com/en/news-articles/more-industries/the-analytical-profile-of-peas
1•mooreds•26m ago•0 comments

Hallucinations in GPT5 – Can models say "I don't know" (June 2025)

https://jobswithgpt.com/blog/llm-eval-hallucinations-t20-cricket/
1•sp1982•27m ago•0 comments

What AI is good for, according to developers

https://github.blog/ai-and-ml/generative-ai/what-ai-is-actually-good-for-according-to-developers/
1•mooreds•27m ago•0 comments

OpenAI might pivot to the "most addictive digital friend" or face extinction

https://twitter.com/lebed2045/status/2020184853271167186
1•lebed2045•28m ago•2 comments

Show HN: Know how your SaaS is doing in 30 seconds

https://anypanel.io
1•dasfelix•28m ago•0 comments

ClawdBot Ordered Me Lunch

https://nickalexander.org/drafts/auto-sandwich.html
3•nick007•29m ago•0 comments
Open in hackernews

Show HN: SiteIQ – LLM and Web security testing tool (built by a high schooler)

https://github.com/sastrophy/siteiq
4•sastrophy•2mo ago
Hi HN! I'm an 11th grade student learning cybersecurity and web development. I built SiteIQ as a hands-on way to understand security vulnerabilities, SEO, and how to test them.

I used AI as my coding partner throughout this project – it helped me understand concepts, debug issues, and write code. Building with AI felt like having a patient tutor available 24/7. I learned way more than I would have just following tutorials.

What it does: - Security Testing: OWASP Top 10 (SQL injection, XSS, CSRF, etc.) - SEO Analysis: Meta tags, schema markup, Core Web Vitals - GEO Testing: Multi-region accessibility and latency - LLM Security: Prompt injection, jailbreaking, system prompt leakage, and "Denial of Wallet" attacks

The LLM security part was the most interesting to build. With everyone adding AI to their apps, I wanted to understand how prompt injection actually works and how to test for it.

Features: - Web UI with real-time console output - CLI for automation - Self-hosted (no data leaves your machine)

Tech: Python, Flask, pytest

GitHub: https://github.com/sastrophy/siteiq

I'd love feedback – are there vulnerabilities I'm missing? Any suggestions for the LLM attack payloads?

This is my first open source project, so any advice is welcome!

Comments

denuoweb•2mo ago
Have you tried running this against itself? I found critical security vulnerabilities:

1. Command Injection Risk (CRITICAL) The web application passes user-controlled input directly to subprocess commands without proper sanitization. An attacker could inject malicious commands through the target_url, wordpress_path, llm_endpoint, or tests parameters. app.py:232-264

2. No Authentication (CRITICAL) All API endpoints are completely unauthenticated. Anyone can start security scans against arbitrary URLs, potentially using your server to attack others. app.py:481-516

3. Server-Side Request Forgery (HIGH) Users can provide any URL as the scan target, allowing attackers to scan internal networks, localhost services, or use your server as a proxy for attacks. app.py:484-493

4. No CSRF Protection (HIGH) POST endpoints lack CSRF token validation, making them vulnerable to cross-site request forgery attacks. app.py:481-482 app.py:567-568

5. No Rate Limiting (MEDIUM) Endpoints lack rate limiting, allowing abuse and denial-of-service attacks.

sastrophy•2mo ago
Thank you so much for taking the time to review the code and pointing these out! I really appreciate it. To clarify: SiteIQ is designed to run locally on your own machine (localhost:5000) as a personal security testing tool - similar to how you'd run Burp Suite or OWASP ZAP locally. It's not meant to be deployed as a public-facing web service. That said, your points are absolutely valid and I'll definitely work on fixing these. This is my first open source project, so feedback like this helps me learn a lot. Thanks again!