Notes
Working notes on agents and LLMs. Drafts, half-written by AI; the numbers in them can still change.
- 75% of LLM Reads Are RedundantI scraped 3 months of AI agent sessions across Claude Code, Codex, and OpenCode. 75% of file reads were redundant.
- A Working-Memory Cache for Agent File ReadsA proposal: let the harness hold one current copy of each active file, instead of storing every read in the conversation.
- What Happened to My Tokens?198 agent spawns ran overnight and half my Codex plan was gone by morning. The culprit was a parent agent that couldn't just wait.