A new Claude Code plugin called TERSE aims to significantly reduce the verbosity of AI model responses, particularly for Anthropic's Fable 5.1 and Opus 5.5 models. The plugin implements a strict set of writing rules designed to cut filler, eliminate jargon, and enforce conciseness, resulting in shorter replies and lower token costs. TERSE can be installed directly within Claude Code or via GitHub, and its rules are integrated into the model's system prompt for consistent application across conversations. AI
IMPACT Could lead to more efficient and cost-effective interactions with large language models by reducing token usage.
RANK_REASON This is a third-party plugin for an existing AI model, not a release from the AI lab itself.
Read on HN — claude cli stories →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →
