Job-search Actors on Apify add 2,139 to 32,737 tokens each to agent tool lists

A developer pinned 20 job-search Actors from the Apify Store one at a time on the hosted Apify MCP server and measured the tool definitions returned by tools/list. The smallest was 2,139 tokens and the median 3,829, while one reached 32,737 tokens, mostly from an output schema the server inferred from Actor results rather than written by the author. Three of the twenty were the author's own, and the measurements were taken on 28 October 2026. The method is a reproducible curl handshake, and token counts used OpenAI's o200k_base tokenizer, so they indicate relative scale rather than exact billing. The findings suggest agents pay a sizable context cost before answering a user's question, driven largely by auto-generated schemas.