Auditing the vocabulary of system prompts and instruction files
Parent: Prompt-side control of LLM lexical tics · Published reference · snapshot 2026-09-23
↓ Facts as markdownall context files
Source-anchored research for Auditing the vocabulary of system prompts and instruction files. Run status: COMPLETED.
These notes link each claim to its source. Some citations identify archived research reports that are no longer hosted on this site. A published reference means the content is available; it does not certify independent review or accuracy.Read the editorial policy and follow the sources before relying on a claim.
How it works
- Anthropic advises that on Opus 4.5/4.6-era models, aggressive emphasis such as 'CRITICAL: You MUST use this tool when...' causes overtriggering; replace with plain 'Use this tool when...'. Audit step: strip shouted/emphatic vocabulary from prompts written for older models. (medium confidence) (verified as of 2026-09-23) [source]
Comparisons and alternatives
- Lexical overuse in LLM output (e.g. 'delve', 'nuanced') is traced to human-feedback training, not prompts: Llama Instruct vs Base shows large increases ('nuanced' +8,342%), and human raters preferred text with these words (52.4% vs 47.6%). A prompt audit cannot remove the underlying bias. (medium confidence) [source]
Facts and statements
- Anthropic docs state prompt formatting style can influence response style; removing markdown from the prompt can reduce markdown in output. Prompt vocabulary and format are therefore an audit surface. (low confidence) (verified as of 2026-09-23) [source]
- Published system-prompt word-ban lists exist (e.g. LLM-Reset lists 'crucial', 'delve', 'matters'); they are practitioner artifacts and their effect is not validated in the papers surveyed. (low confidence) [source]
- Banning a word or phrase in a prompt can suppress a whole cluster of related constructions, so an audit that adds ban lists risks side effects. Source is a practitioner essay, not a controlled study. (low confidence) [source]
Children
- No children recorded.