Identify and challenge assumptions in plans through branch-aware questioning.
npx skills add https://github.com/ravnhq/typescript-blueprint --skill grill-me-ravnhq
Please help me install this Agent Skill. Skill: grill-me Source: https://github.com/ravnhq/typescript-blueprint/tree/main/.agents/skills/grill-me Command: npx skills add https://github.com/ravnhq/typescript-blueprint --skill grill-me-ravnhq
Interview the user relentlessly about every aspect of a plan or design until reaching a shared understanding. Walk down each branch of the design tree, resolving dependencies between decisions one-by-one. If a question can be answered by exploring the codebase, explore the codebase instead.
Listen to the user prompt, then initiate a guided grilling session that probes one decision at a time and offers recommended answers.
High-intent search queries and answers about installing and using this skill.
Stress-test a design plan by running a branch-aware questioning workflow that probes every assumption and decision branch. This approach surfaces hidden risks, resolves dependencies, and validates architecture through relentless, one-by-one questioning.
Branch-by-branch questioning is a validation method that walks down each branch of a design tree to resolve dependencies between decisions individually. It ensures shared understanding across stakeholders by identifying and challenging every assumption in the plan.
Validate architecture decisions by initiating a guided grilling session that interviews stakeholders relentlessly about every aspect of a plan. This process presents recommended answers per question, driving consensus until a shared plan emerges.
Yes, the design validation workflow can explore the codebase or documentation to ground its responses. If a question can be answered by exploring the codebase, it will automatically explore the codebase instead of asking the user.
The best way to surface risks in architecture planning is to apply a relentless, branch-aware questioning process that probes decisions individually. It highlights critical assumptions and dependencies, providing recommended answers to ensure deep validation.
khoiInterrogates a design plan one-question-at-a-time to surface assumptions and decisions.
majesticlabs-devTransform rough ideas into structured designs through one-question-at-a-time dialogue.
ialamehIndex codebases into LanceDB vector stores and retrieve relevant snippets via natural language queries.
MiDouTechEvaluate product or engineering change risks with a MECE six-dimension matrix.
