Unverified50% confidenceTradeoffExact time
Over-specification refers to prompts that hard-code task decomposition steps, tool usage order, and information storage methods, replacing the model's autonomous reasoning with a human cognitive framework, causing the Agent to fail more easily when encountering scenarios not covered.
1
Sources
50%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedOverly verbose, contradictory, or ambiguous system prompts cause the model's attention to become scattered, resulting in over-thinking or instruction-following failures; streamlining agents.md can improve the determinism of instruction following.72% similarUnverifiedOverfitting occurs when a model merely memorizes training data without being able to handle new data, while underfitting occurs when the model is too simple to capture patterns even in the training data.71% similarUnverifiedOver-alignment or over-refusal is a common engineering challenge faced by current mainstream large language models, where models are trained to be overly conservative, resulting in a high false-positive rate for reasonable requests.71% similarUnverifiedOver-engineering refers to introducing complexity into a system design that exceeds current requirements and foreseeable future needs.69% similarUnverifiedExcessive abstraction leads to increased cognitive load, greater debugging difficulty, and accumulated performance overhead.69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/899458API
curl https://kongchang.com/api/v1/knowledge/claims/899458MCP
get_claim(id=899458)