Show HN: Bhowra Ink – C# HTML-to-PDF Under 1 MB, Azure Functions Consumption 4 points · by dillu_spur · 1d · bhowra.com read 1 comment original
Show HN: Turn ad-hoc subagents into durable, accountable AI teams 2 points · by ringlochid · 1d · github.com read comments original
Show HN: Forkbench – the agent gets the API call, not the key (macOS) 2 points · by forkbench · 1d · forkbench.com read comments original
Show HN: Vertex, your personal archiver and annotator 4 points · by asheshambasta · 1d · vertexapp.io read 2 comments original
Show HN: One source of truth for agent configs across repos 3 points · by voidmain42 · 1d · theagencyhq.dev read 1 comment original
Show HN: Knowl – agent memory with write-time supersession, 0.90 on MAB 3 points · by dat999zx · 1d · knowl.cloud read 4 comments original
Show HN: Algolassi – A developer-focused site for .NET, MAUI, Blazor and SQL 2 points · by noobplus · 1d · algolassi.online read comments original
Show HN: 1Presence – AI with memory at scale, layered recall and a vault you own 2 points · by Cloud9Trader · 1d · 1presence.com read comments original
Show HN: Nanointerpret – LLM Interpretability Playground 5 points · by belluxx · 1d · nanointerpret.pages.dev read 1 comment original
Show HN: Timber (iOS) and Timber Tabs (Mac) – best offline read-aloud LLMs 12 points · by Fe2_O3 · 2d · timberreader.com read 1 comment original
Show HN: The lightweight Claude and LMStudio, and memory across sessions. 3 points · by kofdai · 1d · call-me-vera.vercel.app read 3 comments original
Show HN: Millnew AI – Explaining time-series models using local LLMs and Captum 5 points · by ikaru2lt · 1d · robinhood-demo.streamlit.app read 1 comment original
Show HN: Concord – let Claude Code, Codex and Cursor talk to each other 9 points · by alexechoi · 1d · github.com read 4 comments original
Show HN: Flostep – Diagrams people can actually walk through 26 points · by pandurang90 · 3d · flostep.dev read 17 comments original