system prompt — The standing instructions a model receives before any user input — defining its role, rules, tools, and tone for the whole conversation.
AI agent — An AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.
prompt caching — Reusing the model's processed form of a repeated prompt prefix so subsequent calls skip re-reading it, cutting cost and latency substantially.
Why it matters
Understanding these three failure modes helps engineers diagnose perceived 'model drift' in their own coding AI agentAn AI system that doesn't just answer once but works toward a goal in a loop — taking actions, reading the results, and deciding what to do next.Full definition → setups and recognize that reasoning effort settings, prompt cachingReusing the model's processed form of a repeated prompt prefix so subsequent calls skip re-reading it, cutting cost and latency substantially.Full definition → bugs, and system promptThe standing instructions a model receives before any user input — defining its role, rules, tools, and tone for the whole conversation.Full definition → edits can each independently tank output quality.
Key quotes
“We’ve traced these reports to three separate changes that affected Claude Code, the Claude Agent SDK, and Claude Cowork.”
“We never intentionally degrade our models, and we were able to immediately confirm that our API and inference layer were unaffected.”
“A bug caused this to keep happening every turn for the rest of the session instead of just once, which made Claude seem forgetful and repetitive.”
“Because each change affected a different slice of traffic on a different schedule, the aggregate effect looked like broad, inconsistent degradation.”
“All users now default to xhigh effort for Opus 4.7, and high effort for all other models.”
