Interrogate plans one question at a time to resolve design ambiguity.
npx skills add https://github.com/IsKenKenYa/Commory --skill grill-me-iskenkenya
Please help me install this Agent Skill. Skill: grill-me Source: https://github.com/IsKenKenYa/Commory/tree/main/.windsurf/skills/grill-me Command: npx skills add https://github.com/IsKenKenYa/Commory --skill grill-me-iskenkenya
It resolves ambiguity in plans and designs by forcing incremental, decision-tree questioning until both you and the AI reach a shared, well-justified understanding.
Say: “Grill me on my plan to design and implement the next-phase sync architecture, and ask one question at a time until we reach a shared understanding.”
High-intent search queries and answers about installing and using this skill.
To eliminate ambiguity in system design, you can use an iterative questioning approach that traverses each branch of a decision tree. This method stress-tests your plan by asking one question at a time until hidden assumptions and trade-offs become explicit.
Validating an engineering roadmap with branching requirements requires dependency-aware clarification. By resolving dependencies across decisions one-by-one, you ensure later architectural choices do not contradict earlier assumptions and prerequisites become clear.
Stress-testing a local-first vs sync-first architecture plan involves relentless design interrogation. By exploring each decision tree branch incrementally, you force a shared understanding of trade-offs, risks, and prerequisites before implementation begins.
Yes, codebase analysis can answer design review questions instead of guessing. When a question applies to your repository, the system prompts exploration of the codebase to ground architectural decisions in your actual implementation rather than assumptions.
Requirements clarification works effectively for systems thinking and design reviews by applying one-question-at-a-time traversal. This incremental approach forces incremental justification, ensuring both you and the AI reach a shared understanding of complex architectures.
You should not use iterative questioning for architecture decision-making when your plan has no branching dependencies or hidden assumptions. If requirements are already explicit and fully resolved, the one-question-at-a-time traversal provides no additional clarity.
PostalocityScan repository documentation, tokens, and assets to persist design guidelines to a local AI config file.
provenimpactGenerate or update docs/architecture.adoc with C4 diagrams from repository state and feature designs.
rockandrolla13Analyze codebase imports and dependencies to produce a scored architecture diagnostic report.
LeeSandersonInterrogate plans and designs through iterative questioning and decision tree traversal.
