ai on Hacker News

3636 titles from 2320 authors in the last 30 days, with 3358 distinct links and 493 Show HN.

Counts titles, it does not confirm facts. Open the source before concluding anything.

Projects on this term

The score comes from our model.

• Show HN: Make AI coding agents verify its own work · score 8.3

  Reticle is a dev-only SDK and MCP server that lets AI coding agents verify their work by driving the running app and reading network, state, console, and source-level evidence. It returns pass/fail/unknown verdicts with file:line locations, and supports deterministic replayable flows plus CI gating. It addresses the common problem of agents declaring success without checking, filling a clear gap in AI-assisted development workflows.

• Show HN: Loböle, open-source parametric CAD that AI agents can drive · score 8.0

  Loböle is an open-source parametric CAD system built on OpenCASCADE with a Vulkan viewport, exposing a unified command layer accessible from its GUI, Python scripts, and AI agents via JSON-RPC and MCP. It covers sketches, assemblies, FEM simulation, CAM toolpaths, git-backed versioning, and multi-format interchange, and it ships impressive generator-built models like an OM615 diesel and an automatic watch. The agent-driveable architecture is a genuinely novel angle for desktop CAD.

• Show HN: Keep your credentials out of the AI agents you use and build · score 7.8

  Weir is a security tool that detects when AI agents leak credentials by planting unique fake keys in the agent's environment, running its flows, and checking every output channel (spans, logs, prompts, network) for those exact values. It also integrates with Claude Code to monitor and redact real credentials during sessions. Notably, it was used to discover a real vulnerability in MLflow.

• Show HN: Hackerscad – Code-first CAD with MCP server for AI agents · score 7.8

  Hackerscad is a code-first parametric CAD tool where models are written in a small D-like scripting language, with mouse interaction that writes code back, and an MCP server enabling AI agents to create and manipulate 3D models. It combines OpenCascade B-rep solids, a custom language with sandboxing, bidirectional code/GUI editing, and AI-agent integration into a single desktop application. The project is technically ambitious and fills a gap for AI-assisted parametric CAD.

Recent titles

1. Beyond the Breakthrough: What Comes Next for AI [video] [Peter Norvig]
   2026-11-01 · youtube.com · HN discussion
2. DIY AI Lockpicking Robot: ESP32 and Claude AI Opens Any Lock [video]
   2026-11-01 · youtube.com · HN discussion
3. AI Software Clones Are Coming for Big-Name Companies Like Adobe
   2026-11-01 · engadget.com · HN discussion
4. AI Data Annotation and Evaluation Contractor
   2026-11-01 · iungotech.com · HN discussion
5. Measuring AI Impact: What Your "Velocity Chart" Cannot Tell the CFO
   2026-11-01 · age-of-product.com · HN discussion
6. I Review 2k Lines of AI Generated Code
   2026-11-01 · bytesizedbets.com · HN discussion
7. Ask HN: Is last week a signal that AI is a bubble waiting to burst
   2026-11-01 · HN discussion
8. Ransomware gang bypassed AI guardrails by posing as red-team testers
   2026-11-01 · sophos.com · HN discussion
9. How to Prove What Your AI Agent Did – ZizkaDB's Two GitHub Actions
   2026-11-01 · medium.com · HN discussion
10. AI as Social Technology
    2026-11-01 · knightcolumbia.org · HN discussion
11. Opt out of AI use at work based on religious freedoms law
    2026-11-01 · fortune.com · HN discussion
12. Terence Tao on how AI math solutions miss the point
    2026-11-01 · sciencefriday.com · HN discussion
13. Show HN: Let formal specs test AI implementations
    2026-11-01 · github.com · HN discussion
14. AI agent swarms in Minecraft fail to build civilization, drown in bureaucracy
    2026-11-01 · tomshardware.com · HN discussion
15. What does good AI mean?
    2026-11-01 · thoughts.wyounas.com · HN discussion
16. AI is a technological version of the Homeric songs
    2026-11-01 · economist.com · HN discussion
17. Vibe Coding an Autonomous Agent Harness in Go: Self-Healing AI Workflows
    2026-11-01 · medium.com · HN discussion
18. Agentic AI is myth set in motion
    2026-11-01 · programmablemutter.com · HN discussion
19. A 90% Chance of AI Saving Humanity
    2026-11-01 · joshbocanegra.medium.com · HN discussion
20. The question we should all be asking about AI
    2026-11-01 · cnn.com · HN discussion
21. Context Mode – An Open-Source MCP Server That Reduces AI Coding Context Usage
    2026-11-01 · github.com · HN discussion
22. When AI Writes the Code, Show Your Judgment
    2026-11-01 · eyosias.dev · HN discussion
23. Will Agentic Resource Registry ever have its moment, the way AI gateways did?
    2026-11-01 · agenticresourcediscovery.org · HN discussion
24. Show HN: AIOps Cockpit – An open-source AI SRE with human-in-the-loop safety
    2026-11-01 · github.com · HN discussion
25. AI Warning for Steam – Increases Visibility of AI Generated Content Disclosures
    2026-11-01 · github.com · HN discussion
26. Fruth – A semantic state inference engine for AI
    2026-11-01 · github.com · HN discussion
27. FreeGent: Free Agentic AI for Everyone
    2026-11-01 · freegent.ai · HN discussion
28. What operational task would you hand to a 7-day custom AI tool?
    2026-11-01 · indiehackers.com · HN discussion
29. Early Career Researcher Responses to AI in Math
    2026-11-01 · arxiv.org · HN discussion
30. Is AI Cheap Enough to Replace Developers?
    2026-11-01 · zyvop.com · HN discussion

All trends