The KLOD blog https://keepthelightson.dev/blog/ Engineering, people and keeping essential work possible when AI stops. en What the KLOD check found in precommitEU https://keepthelightson.dev/blog/2026/09/what-the-klod-check-found-in-precommiteu/ I ran the klod-check skill on my own compliance scanner before asking anyone else to. Here is the report, what it got right, and the one gap it found. Case study Sat, 12 Sep 2026 00:00:00 GMT https://keepthelightson.dev/blog/2026/09/what-the-klod-check-found-in-precommiteu/ Vibe coding is fine. Then the engineering starts. https://keepthelightson.dev/blog/2026/09/beyond-vibe-coding/ AI can help you build a prototype quickly. Taking it into production still means checking permissions, handling failures and giving people a way to take over. Engineering judgement Fri, 11 Sep 2026 00:00:00 GMT https://keepthelightson.dev/blog/2026/09/beyond-vibe-coding/ When AI stops, can your team take over? https://keepthelightson.dev/blog/2026/09/keep-the-people-who-keep-you-running/ Your team needs time to learn, tools they can use and practice handling real work without AI. Here's how to make that part of the job. People and continuity Fri, 11 Sep 2026 00:00:00 GMT https://keepthelightson.dev/blog/2026/09/keep-the-people-who-keep-you-running/ Adding KLOD to Spec Kit and OpenSpec https://keepthelightson.dev/blog/2026/09/klod-with-spec-kit-and-openspec/ Where the two KLOD steps fit in a Spec Kit or OpenSpec workflow, and the exact commands to install them from the public repository. Tools and workflow Fri, 11 Sep 2026 00:00:00 GMT https://keepthelightson.dev/blog/2026/09/klod-with-spec-kit-and-openspec/ Hello. This is KLOD. https://keepthelightson.dev/blog/2026/09/welcome-to-klod/ I'm Alex. KLOD is about building useful AI software while making sure people can keep essential work going when the model is unavailable. Start here Fri, 11 Sep 2026 00:00:00 GMT https://keepthelightson.dev/blog/2026/09/welcome-to-klod/ If the model starts working against you, can you switch it off? https://keepthelightson.dev/blog/2026/09/when-ai-breaks-trust/ What AI safety research means for everyday software: limit an agent's permissions, keep an independent stop control and make unfinished work easy to take over. Trust and control Fri, 11 Sep 2026 00:00:00 GMT https://keepthelightson.dev/blog/2026/09/when-ai-breaks-trust/