In useContinuously updated · agent evals not yet graded
My personal tooling: an agent bench of Claude skill specs with explicit refusal conditions, plus an Obsidian second brain.
- Claude Code Skills (SKILL.md)
- Claude API · Managed Agents
- Python
Personal tooling
Systems
JonnyJr, the infrastructure I built for my own work: an agent bench and an Obsidian second brain. Not a product — it’s judged on whether it changes how I work.
Every project page follows the same template: problem, requirements, architecture, design decisions, scope, implementation, verification, failures, limitations, and links. Sections without evidence yet are marked pending rather than filled in.