6 posts across articles, notes and newsletter issues.
Three agents on one checkout means trampled files and lost work. Giving each task its own git worktree, its own model and a deterministic merge step fixed that for me.
Clone, install, run the tests, then separate real regressions from environmental noise. Why I built repo-doctor as a deterministic pipeline first and an agent second.
A short note from building cron-for-agents. The first version kept its watermark in memory, a restart erased it, and an agent reprocessed a week of old data. State belongs on disk.
Plain cron runs your agent on time, then wastes tokens reprocessing old data and spams you with identical summaries. Watermarks and content-hash dedup fix both.
Two months of unglamorous infrastructure. A cron tool that does not spam duplicate summaries, sandbox M-Pesa tools for agents, and the start of a proper way to test how agents fail.
Worktrees do not disappear when an agent's task ends. A short note on the cleanup step I almost skipped while building worktree-orchestrator, and the disk space and confusion it saves.
