Quota burn and style-guide ignoring are the same attention bug, and how to fix it
RandalSchwartz · reddit · 2026-10-30
A Reddit deep-dive argues that "my agent burns my 7-day quota in 3 days" and "my agent ignores my style guide" are one closed loop: pasting bad code under "DON'T do this" loads those tokens into the KV-cache, where they act as semantic attractors and prime the model to emit exactly the nullable unions and defensive fallbacks you banned — then 25 turns of diff-pasting at 50k tokens each drains your quota. Fixes validated across 238 production tickets: replace negative examples with a 4-part override structure (name the crash → call out the tutorial shortcut → permit the agent to halt and counter-offer → give the positive sequence), and pass call-graph blast radii between subagents using Graphviz strict digraph instead of prose or Mermaid, which taps the model's LLVM/GCC compiler control-flow training.
More from coding & agent
- Vercel quietly launches "Agentic Infrastructure", positioning itself as the deploy target for coding agents — evilrabbit_ · 2026-10-30
- Garry Tan shows one-command GBrain: hosted Postgres+pgvector memory for any coding agent — garrytan · 2026-10-30
- Mollick on Opus 5.5: prompt less, build in checkpoints instead of one-shotting — eldonredwards · 2026-10-30
- Hugging Face Spaces' Agents button lets your AI agent use any demo for free — victormustar · 2026-10-30
- Atelico launches fully on-device game AI engine with LLMs, voice, and NPC behavior — nathanbenaich · 2026-10-30
- Hooking Kimi up to bloodwork via MCP: a Redditor mined years of lab results in one evening — Local-Parking5561 · 2026-10-30