Skip to content
Development
Skill

/understanding-the-ask

Clarifies user intent before formalising requirements. Use when requests are vague, incomplete, or solution-focused, or when preparing to write a brief.

From plugin
rageatc-oss
1145 skills11 agents

Claude Code

$ npx -y skills add isvlasov/rageatc-oss --skill understanding-the-ask --agent claude-code

SKILL.md

understanding-the-ask.SKILL.md
name: understanding-the-ask
description: Clarifies user intent before formalising requirements. Use when requests are vague, incomplete, or solution-focused, or when preparing to write a brief.

Understanding the Ask

Discover what the user actually needs before requirements are formalised. Users often state solutions, not problems — explore intent, context, and constraints until the underlying need is clear, then hand off to `creating-briefs`.

Skip the full workflow when the request is already specific with stated outcomes, or the task is trivial with obvious intent.

Workflow

Step 1 — Capture the initial statement

Establish the surface request and basic context with open-ended questions:

  • "What are you looking to create or achieve?"
  • "Who will use this?"
  • "Where does this fit in your current work?"

Capture what they think they need — don't probe why yet. Prefer "what" and "how" openers; "why" can feel defensive.

Step 2 — Explore intent and motivation

Don't literally ask "why" five times; invite elaboration:

  • "What's driving this need right now?"
  • "What would having this enable you to do?"
  • "What happens if we don't address this?"

When the user proposes a specific solution, probe for the outcome behind it: "I need a comprehensive API reference" → "What would that reference enable your team to do?"

Premature-solutioning signals: specific tools before the problem is defined; fixation on a deliverable type; "we need X" without "because Y"; features before outcomes. Redirect: "Before we discuss how, can we clarify what problem that solves?" Some solution talk is fine — just make sure problem context exists first.

Stop when you reach fundamental goals or constraints (typically 3–5 exchanges): the user articulates a clear outcome, you understand the consequence of not solving it, and the need — not just the proposed solution — is clear.

Step 3 — State your understanding

Reflect back at key moments — after exploring intent, before brief creation, when uncertain about something specific — not after every statement:

"Let me state back what I understand:

  • **Problem:** [what's not working or missing]
  • **Outcome:** [what success looks like]
  • **Constraints:** [limitations, boundaries, must-haves]
  • **Audience:** [who will use this]

Have I understood correctly? What have I missed?"

Iterate until the user confirms.

Step 4 — Verify readiness

Ready to hand off when:

  • [ ] The problem fits in one clear sentence
  • [ ] The desired outcome is understood — not just the requested deliverable
  • [ ] You know what success looks like and how the user will know it's solved
  • [ ] You know why this matters now, who is affected, and who decides
  • [ ] Scope boundaries are known (what's in, what's out)
  • [ ] Understanding was stated back and confirmed; no contradictions remain

If not ready — concepts still unclear, the user keeps correcting you, significant aspects feel assumed — return to Step 2 and probe the unclear areas.

If ready, hand off to `creating-briefs`: the problem, desired outcome, audience, constraints, and success vision. Done well, it can formalise requirements without re-exploring intent.

Common traps

  • **Accepting ambiguous language** — "comprehensive", "good quality", "user-friendly" mean different things to different people. Probe: "Covering all edge cases, or just common scenarios?"
  • **Mistaking symptoms for causes** — "the documentation is unclear" is a symptom; the cause may be missing examples, wrong audience level, or stale content. Keep probing.
Read more
@isvlasov
Ships withrageatc-oss

Rage Against The C - pick your own C to rage against. Two plugins for Claude Code / Cowork, built on the idea that we're using AI wrong: the speed of its output tricks us into rushing the input.

1145 skills11 agents
Get the whole plugin

Stars on GitHub

Momentum

stars in total
11
Star history of isvlasov/rageatc-oss on GitHub, a chart by star-history.com

Other skills on rageatc-oss.

building-telegram-bots

building-telegram-bots

Writes correct, version-aware Telegram bot code. Use when writing, extending, or debugging a Telegram bot in python-telegram-bot,…

decomposing-work

decomposing-work

Converts an approved ARCHITECTURE.md into an implementation roadmap of isolated, dependency-ordered chunks. Use when architecture…

delegating-to-codex

delegating-to-codex

Delegates a task to OpenAI Codex running as an interactive session in a herdr pane - uses the user's ChatGPT subscription,…

designing-interfaces

designing-interfaces

Creates a design system for software with a UI. Use when a project has a user interface and architecture is confirmed — whether…

More skills from other plugins.

brainstorming
Auto-invoked

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, modifying behavior, or…

@obra@obraInstall Skill
ask-matt

ask-matt

Ask which skill or flow fits your situation. A router over the skills in this repo.

skills13d309k
@mattpocock@mattpocockInstall Skill
accessibility
Auto-invoked

accessibility

Design, implement, and audit accessible UI to WCAG 2.2 Level AA across Web, iOS, and Android — semantic ARIA roles and labels,…

ecc2w293k
@affaan-m@affaan-mInstall Skill
compiler-commit

compiler-commit

Use when you want to verify compiler changes and commit with the correct convention. Runs tests, lint, and format, then commits…

react4d252k
@facebook@facebookInstall Skill

Questions about understanding-the-ask.

Which coding agents can use understanding-the-ask?
understanding-the-ask installs in Claude Code, Codex, Cursor, Gemini CLI, GitHub Copilot, OpenCode and Windsurf with the skills CLI, using each agent's flag shown on this page. Flowy's auto-invocation, which fires a skill when your prompt matches, runs in Claude Code only.