EBEmerson Brauninagentskit.hashnode.dev·8h ago · 6 min readI stopped asking coding agents to guess where they could editMost mistakes I see from coding agents do not begin with bad code. They begin one step earlier. The agent opens the wrong documentation, assumes a neighboring package owns the change, edits a director00
EBEmerson Brauninagentskit.hashnode.dev·9h ago · 6 min readI stopped writing rules for coding agents that CI could not enforceI used to treat repository instructions as the finished product. Write a careful AGENTS.md. Explain the architecture. Ban any. Require named exports. Tell the coding agent to keep files small and upda00
EBEmerson Brauninagentskit.hashnode.dev·9h ago · 6 min readI stopped installing opaque AI agents and started copying their sourceAn agent catalog can look useful right up until the agent enters your repository. The card says “research agent.” The install button says “ready.” What I actually need to know is less exciting: Which00
EBEmerson Brauninagentskit.hashnode.dev·21h ago · 6 min readI removed an arbitrary-SQL RPC from an agent memory adapterThe first version worked. That was the problem. I was adding Supabase-hosted pgvector support to an agent memory adapter. The fastest path was tempting: send SQL to a database function, let the functi00
EBEmerson Brauninagentskit.hashnode.dev·5d ago · 7 min readThe model call was the easy part: keeping an agent stack replaceableGetting an agent to answer was the easy part. The architecture became difficult later. Model-specific types started crossing boundaries. Tool calls depended on one SDK. Chat state lived inside UI comp00
