Show HN ranking for the past week

  1. boringBar – a taskbar-style dock replacement for macOS | 305
  2. Smol machines – subsecond coldstart, portable virtual machines | 93
  3. Oberon System 3 runs natively on Raspberry Pi 3 (with ready SD card) | 111
  4. I built a social media management tool in 3 weeks with Claude and Codex | 131
  5. Every CEO and CFO change at US public companies, live from SEC | 64
  6. PanicLock – Close your MacBook lid disable TouchID –> password unlock | 69
  7. Ithihāsas – a character explorer for Hindu epics, built in a few hours | 45
  8. MacMind – A transformer neural network in HyperCard on a 1989 Macintosh | 42
  9. LangAlpha – what if Claude Code was built for Wall Street? | 54
  10. Libretto – Making AI browser automations deterministic | 54
  11. Claudraband – Claude Code for the Power User | 44
  12. SPICE simulation → oscilloscope → verification with Claude Code | 30
  13. Stage – Putting humans back in control of code review | 96
  14. CodeBurn – Analyze Claude Code token usage by task | 24
  15. I made a calculator that works over disjoint sets of intervals | 8
  16. Plain – The full-stack Python framework designed for humans and agents | 51
  17. Marky – A lightweight Markdown viewer for agentic coding | 35
  18. Hormuz Havoc, a satirical game that got overrun by AI bots in 24 hours | 17
  19. Home Memory – A local DB of my house, down to cables and pipes | 12
  20. A memory database that forgets, consolidates, and detects contradiction | 33
  21. Kelet – Root Cause Analysis agent for your LLM apps | 24
  22. Waffle – Native macOS terminal that auto-tiles sessions into a grid | 23
  23. I rebuilt a 2000s browser strategy game on Cloudflare's edge | 33
  24. Hiraeth – AWS Emulator | 8
  25. Gave Claude a casino bankroll – it gambles till it's too broke to think | 9
  26. Fakecloud – Free, open-source AWS emulator | 9
  27. SmallDocs - Markdown without the frustrations | 20
  28. Continual Learning with .md | 34
  29. Ilha – a UI library that fits in an AI context window | 9
  30. Avec – iOS email app that lets you handle your Gmail inbox in seconds | 4
  31. Make sure your OpenClaw isn't doing things it's not supposed to | 3
  32. Omi – watches your screen, hears conversations, tells you what to do | 13
  33. Agent-cache – Multi-tier LLM/tool/session caching for Valkey and Redis | 6
  34. Revdiff – TUI diff reviewer with inline annotations for AI agents | 4
  35. A CLI that writes its own integration code | 10
  36. IceGate – Observability data lake engine | 5
  37. A stateful UI runtime for reactive web apps in Go | 4
  38. Equirect – a Rust VR video player | 1
  39. Object Storage Comparisons | 0
  40. Mcptube – Karpathy's LLM Wiki idea applied to YouTube videos | 2
  41. Editing 2000 photos made me build a macOS bulk photo editor | 25
  42. SmokedMeat, like Metasploit, but for CI/CD (open-source) | 9
  43. Host, share, promote and track release artifacts via your coding agent | 0
  44. Pushduck – S3 uploads that run on Cloudflare Workers, no AWS SDK | 7
  45. We built an AI Agent to reproduce bugs | 2
  46. Jeeves – TUI for browsing and resuming AI agent sessions | 3
  47. Git why – log your agent reasoning trace along your code | 1
  48. Compile English specs into 22 MB neural functions that run locally | 0
  49. Irl.rent – What renters pay, not what's listed in SF | 2
  50. A Better Internet | 3
  51. A social feed with no strangers | 5
  52. React Email 6.0 – An open-source email editor you can embed in your app | 0
  53. Xit – a Git-compatible VCS written in Zig | 1
  54. I built a Wikipedia based AI deduction game | 7
  55. ParseBench – Document parsing benchmark for AI agents | 5
  56. AI support chatbot with RAG and citations – one back end file, no infra | 0
  57. Rekal – Long-term memory for LLMs in a single SQLite file | 10
  58. Voiden – API Workflows as Markdown Files (Obsidian × Curl) | 5
  59. Deflect One – command line dashboard for managing Linux servers via SSH | 7
  60. I benchmarked Gemma 4 E2B – the 2B model beat the 12B on multi-turn | 1
  61. Lazyagent – a local TUI for watching what your coding agents are doing | 0
  62. OQP – A verification protocol for AI agents | 3
  63. Lazyagent – TUI for to watch all your AI coding agents | 0
  64. German Abitur: More good grades, fewer bad grades | 3
  65. Slack killed their OpenAPI spec, so we reverse engineered it | 1
  66. We built an MCP for Windows – ask Claude about CPU, temps, and privacy | 5
  67. Skillgrab – scan any project, auto-install matching AI skills | 1
  68. XitDB – an immutable single-file database | 0
  69. Dbg – One CLI debugger for every language (AI-agent ready) | 0
  70. We scanned uscis.gov for third-party trackers. The results are jarring | 0
  71. US keyboards don't have enough keys, so I switched to Japanese | 0
  72. Tsplat – Render Gaussian Splats directly in your terminal | 1
  73. I'm organizing a vibe coding game dev competition | 0
  74. CJIT, a single-binary C compiler that can self host | 1
  75. pg_grpc – Call gRPC services directly from PostgreSQL | 1
  76. Agent Armor, a Rust runtime for enforcing policies on AI agent actions | 5
  77. T4 – a versioned datastore with branching and time-travel (S3-backed) | 2
  78. Bloomberg Terminal for LLM ops – free and open source | 1
  79. One-command deploys from your terminal with CreateOS CLI | 1
  80. Keyboard First Email Client | 2
  81. Sudomake Friends, personalized AI personas in a Telegram group chat | 1
  82. OpenRig – agent harness that runs Claude Code and Codex as one system | 5
  83. StockFit API – structured SEC EDGAR data with a free tier | 1
  84. Noodlist – Letterboxd for Instant Ramen | 9
  85. Open-source Perplexity clone one file back end, streaming answers | 1
  86. Flint – A 30B model fine-tuned for less repetition | 2
  87. Agent Citizen – Your AI agents are sitting around doing nothing | 1
  88. Chrome extension that extracts styles and generates DESIGN.md files | 2
  89. Moxn – Git-like version control for collaborative docs | 7
  90. Turn your favorite YouTube channels into a streaming experience | 10
  91. Pyra – a Python toolchain experiment inspired by uv and Bun | 1
  92. Run AI coding agents in real, local sandboxes, not Git worktrees | 3
  93. Git-stage-batch – curate clean commit history before pushing | 1
  94. Terminal-Wrench, a dataset of 331 realistic hackable environments | 2
  95. Uninum – All elementary functions from a single operator, in Python | 2
  96. Dependicus, a dashboard for your monorepo's dependencies | 1
  97. Mercury – No-code orchestration for human and agent teams | 5
  98. BuilderPulse – Daily intelligence from 10 sources for indie hackers | 4
  99. Open KB: Open LLM Knowledge Base | 2
  100. MCP server gives your agent a budget (save tokens, get smarter results) | 5