1. Show HN: TurnKeeper – Fair, explainable team rotations for Slack (no AI) (getturnkeeper.com)
Whose turn is it? Fair, explainable rotations for Slack teams. No spreadsheets, no AI guesswork. €19/month flat.
2. Show HN: Termaxa – my agent gate passed its security rig, failed a two-user test (github.com)
A cooperative gate for the shell commands AI agents run. Previews, backups, policy, audit. Claude Code + Cursor. A windshield, not a sandbox. - termaxa/termaxa
3. Show HN: I benchmarked LLMs on predicting knife steel properties (github.com)
Open benchmark: how well can LLMs predict knife-steel properties (edge retention, toughness) from chemical composition, scored against CATRA/Charpy lab measurements. - Steel-predictor-project/steel...
4. Show HN: Stunt – a stunt double for the APIs you integrate (95 adapters) (github.com)
Local simulators for public APIs — mock/stub servers for testing without remote accounts, network, or bills. Go + sandboxed Starlark; 95 adapters embedded. - StuntAPI/stunt
5. Show HN: Europe Robotics Map – An open source map of robotics companies (europeroboticsmap.com)
An open, community-maintained atlas of robotics and robotics-adjacent companies, offices and hiring locations across Europe.
6. Show HN: PhysiClaw – an AI agent that physically operates an iPhone (github.com)
The AI agent that interacts with you in the real world. - physiclaw/PhysiClaw
7. Show HN: A searchable index of GitHub's lists that shows what's dead (awesome.donld.me)
32.2K projects from 80 curated awesome lists, sorted by stars and kept under the categories their curators wrote, each showing how long since anyone touched it.
8. Show HN: Convex clients in 100 programming languages (github.com)
Experimental Convex clients across 100 programming languages, built and verified in Docker. - mikecann/the-convex-100-language-challenge
9. Show HN: TextCompare – Local-first diff tools for text, files, and code (compare-text.org)
Compare two texts online and see every difference highlighted instantly. Line, word and character diff, side-by-side or inline view — free, fast and 100% private: your text never leaves your browser.
10. Show HN: A benchmark for AI agent guardrails that caught my own plugin (github.com)
Holdline — a neutral benchmark for AI-agent write-guards. Catch rate, false-block rate, class-balanced kappa, and an injection-attack class. Scores any guard. - couldbeme/holdline