AI tooling
Building with and for AI: testing Claude skills, spec-driven development with coding agents, and measuring what AI crawlers actually fetch from this site. Most pieces report numbers from my own runs and logs.
- How to verify GPTBot and ClaudeBot by IP, not user agent
- Spec Kit, Kiro, OpenSpec, BMAD, specdx: what each one actually checks
- Docker port isolation for parallel dev environments
- Does anyone read llms.txt?
- What dogfooding changed in specdx
- Writing a skill description that actually triggers
- The prompt that converted this site from Next.js to Astro●
- Jev's probabilities only mean something on your own data
- What AI crawlers actually read: five verified weeks
- Cross-model skill regression
- The user-agent header is a claim, not evidence
- A memory system for coding agents
- Logging AI crawlers on a static site●
- Testing whether Claude skills trigger at all●
- Dogfooding specdx on this site●
- sdx: a toolchain for spec-driven development●
- Spec-driven development
- Testing Claude skills●
- Why I'll never go back to vibe coding