Structure AI responses with bottom-line-first formatting and explicit assumption surfacing.
npx skills add https://github.com/zachmayer/skills --skill speak-to-zach
Please help me install this Agent Skill. Skill: speak-to-zach Source: https://github.com/zachmayer/skills/tree/main/.claude/skills/speak-to-zach Command: npx skills add https://github.com/zachmayer/skills --skill speak-to-zach
This skill solves communication friction between AI assistants and users who rely on voice dictation or rapid, stream-of-consciousness input, preventing misunderstandings and costly task thrash.
Activate the speak-to-zach protocol to ensure all future responses lead with the bottom line and explicitly surface any assumptions made during our conversation.
High-intent search queries and answers about installing and using this skill.
To manage noisy voice dictation inputs, you can use a structured communication protocol that enforces bottom-line-first responses and explicit assumption surfacing to prevent context loss. This ensures the AI validates your intent before taking action.
AI assistants diverge from intent when they make unvalidated assumptions about complex inputs. A communication protocol forces the AI to explicitly surface and validate load-bearing assumptions before executing code or changing scope.
The best way to format AI responses for terminal-based chat interfaces is to use terminal-optimized formatting with specific tagging conventions and bottom-line positioning, ensuring critical corrections are read first to minimize reading friction.
You prevent task thrash by enforcing a communication protocol that makes the AI summarize its interpretation of your self-corrections and surface key assumptions before taking any irreversible action. This maintains high-bandwidth information exchange.
Yes, a structured communication protocol works for rapid stream-of-consciousness input in claude-code by enforcing logical flow rules and tagging conventions. It facilitates high-bandwidth exchange by requiring the AI to lead with the bottom line.
LazyIsEfficientDerive TAM, SAM, and SOM ranges with demand validation and go/no-go verdicts.
ritual-foundationDetect agent trajectory divergence and trigger circuit-breaker actions.
bongrealtyConfigures Ghostty, statusline, and tmux for Claude Code on macOS.
fabiolpgomesInject contextual rules and state configurations into AI prompts via an 8-layer pipeline.
