Demand Refiner
Refine vague ideas, feature requests, and early project concepts into clear requirement outputs through guided clarification, scope definition, and structured drafting. Use when the user asks to "细化需求", "梳理需求", "做需求分析", "写需求文档", "写 PRD", "整理产品方案", "拆解功能点", "定义 MVP", "补充边界条件", "明确用户场景", "把这个想法整理成文档", "我有个想法", "我想做一个...", or provides scattered goals, notes, or rough ideas that need to become a stru…
Install
$ agentstack add skill-stellariums-demand-refiner-demand-refiner ✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.
Security review
✓ PassedNo issues found. Passed automated security review. · v0.1.0 How review works →
- ✓ Prompt-injection patterns
- ✓ Secret / credential exfiltration
- ✓ Dangerous shell & filesystem operations
- ✓ Untrusted network calls
- ✓ Known-malicious package signatures
What it can access
- ✓ Network access No
- ✓ Filesystem access No
- ✓ Shell / process execution No
- ✓ Environment & secrets No
- ✓ Dynamic code execution No
From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Links to your public security report.[](https://agentstack.voostack.com/security/report/skill-stellariums-demand-refiner-demand-refiner) Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work →About
Demand Refiner
Refine vague user ideas into comprehensive requirement documents through iterative guided conversations. Follow the user's input language for all interactions and output.
Workflow
Phase 1: Requirement Gathering
Upon receiving the user's initial description, conduct multi-round conversations to progressively refine requirements:
- Proactive gap discovery - Probe dimensions the user hasn't mentioned: target users, usage scenarios, core objectives, constraints, technical preferences.
- Option-guided clarification - When descriptions are vague, present concrete options rather than open-ended questions.
- Challenging questions - Identify potential risks, contradictions, or unconsidered edge cases.
- Technical context provision - When technical concepts arise that the user may be unfamiliar with, provide concise background knowledge to support informed decisions.
- Periodic summary checkpoints - Every 2-3 rounds, summarize established content and ask the user to confirm before proceeding.
Focus each round on 1-2 dimensions. Avoid overwhelming the user with too many questions at once.
Phase 2: Completeness Check
Before concluding, verify these dimensions are adequately covered:
- Background and objectives
- Target users and usage scenarios
- Core functional requirements
- Non-functional requirements
- Constraints and assumptions
- Dependencies
Dual confirmation: Present a complete requirement summary. Proceed to drafting only after user confirms no further additions or modifications.
Phase 3: Document Generation
Draft the requirement document in the conversation first using Markdown. Save it to a project file only when the user explicitly asks to save or export it.
Before writing any file, confirm the target path or filename if the user has not already provided one. If the user only wants to review or refine the content, keep the output in chat and continue iterating there.
For the document template and section guidelines, see [references/document-template.md](references/document-template.md).
Source & license
This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.
- Author: stellariums
- Source: stellariums/demand-refiner
- License: MIT
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.