I Built a Tool Because 90% of My AI Agent's Tokens Were Spent Searching, Not Coding

5,000 to 10,000 tokens. That's what my AI agent was burning before changing a single line of frontend code. I started paying attention to this a while back. I'd ask the agent to do something small ("fix the padding on that card," "swap this to the secondary variant") and the actual edit would be maybe 200 tokens of real work. But the agent wouldn't start with the edit. It would start with a search. Grepping the codebase, reading five files, asking me "is this the right component?", reading files, building context. All of that before it touched a line of code.