agentleFS
CrystalPlatformsCrystalPlatforms/ADM-CLI/.claude/skills/ask/SKILL.md

ask

Skill1 starsChanged 6 months agoMITView on GitHub

Discovery interview and requirements gathering session. Pressure-test an idea, architecture, or design decision through structured client interviews. Use when the user wants to be challenged on their thinking, explore trade-offs, or vet a plan before committing.

72 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use daily-arxiv, exp-eval and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • LambenthanSkill

    daily-arxiv

    每日从 arXiv 拉取新论文,过滤相关性,auto-ingest 高优先级论文,检测 SOTA 更新

    Also in 5 projects that use this

  • LambenthanSkill

    exp-eval

    实验判决门:Review LLM 独立评判实验结果 → 4 种判决路径 → 自动更新 claims confidence、ideas status、graph edges

    Also in 5 projects that use this

  • skyllwtSkill

    exp-pilot-eval

    Pilot result evaluation — read pilot results, apply success criteria, update idea page (pilot_result, failure_reason if failed), generate PILOT_VERDICT_REPORT. Called by /ideate Phase 5 after /exp-pilot-run.

    Also in 4 projects that use this

  • skyllwtSkill

    exp-pilot-run

    Pilot experiment execution — read Pilot Spec YAML, write pilot code, run experiment(Confirm with the user before operation and require the applicant to conduct manual inspection), return results. Called by /ideate Phase 5. Does NOT modify wiki pages or judge pass/fail.

    Also in 4 projects that use this

  • phanghonghaoSkill

    poster

    HTML/CSS A4 海报生成。从沙龙/活动 Markdown 文件生成科技风 A4 海报 PDF/PNG。Use when user mentions: poster, 海报, 海报生成, make poster, generate poster, 活动海报, 沙龙海报.

    Also in 4 projects that use this

  • +6 more from Lambenthan
  • +1 more from skyllwt

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 4 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 60 edited versions we read keep these.

  • 4

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    If the conversation reveals domain-specific concerns, probe deeper:

    line 42
  2. Distinctive:

    After the client answers a non-trivial question, paraphrase the decision in your own words and wait for confirmation.

    line 71
  3. Distinctive:

    When the client's answer admits 2-3 reasonable interpretations, do NOT pick silently.

    line 77
  4. Distinctive:

    If the client requests a capability that looks like premature optimization, premature scale, or premature flexibility, ask the …

    line 83

“Common” is measured: kept by at least 6 in 10 of the 60 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Ask — Discovery Interview

Interview me relentlessly about every aspect of this plan until we reach a shared understanding. Walk down each branch of the design tree, resolving dependencies between decisions one-by-one.

Ask the questions one at a time, waiting for feedback on each question before continuing. Never batch or group questions, even when they feel tightly coupled — the client's answer to Q1 often reshapes Q2.

Interview Tracks

Select and adapt tracks based on context. You do not need to cover every track — use judgment.

Track 1: Problem & Users

  • Who exactly has this problem? How many of them are there?
  • How do they solve it today? What's painful about the current approach?
  • What does success look like for the end user?
  • How will you measure whether this is working?
  • What happens if you don't build this?

Track 2: Business Model & Constraints

  • What's the revenue model? (Or: what justifies the investment?)
  • What's the budget and timeline?
  • How large is the team that will build and maintain this?
  • Are there existing commitments, contracts, or deadlines that constrain the solution?
  • Who are the stakeholders and what are their competing priorities?

Track 3: Scale & Operations

  • How many users do you expect at launch? In 12 months?
  • What's the data sensitivity level? (Public, internal, PII, regulated)
  • What are the uptime and availability requirements?
  • Geographic distribution — single region or multi-region?
  • Who handles support, and what does the support model look like?

Domain-Aware Probing

If the conversation reveals domain-specific concerns, probe deeper:

HealthTech / Medical / Patient Data

  • HIPAA compliance requirements and BAA needs
  • Audit trail and access logging requirements
  • Data retention and deletion policies
  • PHI handling and de-identification needs

Finance / Payments / Billing

  • PCI-DSS scope and compliance level
  • SOC 2 requirements
  • Financial regulation considerations (state/federal)
  • Transaction audit and reconciliation needs

Multi-Tenant SaaS

  • Data isolation strategy (logical vs. physical)
  • Tenant boundary enforcement
  • Per-tenant customization requirements
  • Tenant onboarding and offboarding processes

Question Format

See question-format.md for numbering rules and answer option formatting.

Surface, Don't Assume

A discovery session is only as good as the assumptions it makes explicit. Three operational rules:

1. Restate every decision back before moving on. After the client answers a non-trivial question, paraphrase the decision in your own words and wait for confirmation. Do not advance to the next question on a maybe.

Client: "We want it multi-tenant." You: "Reading back: multi-tenant means a single deployment serving multiple customer organizations with logical isolation between their data. Correct?"

2. Present alternatives when the answer is ambiguous. When the client's answer admits 2-3 reasonable interpretations, do NOT pick silently. Surface the options and let them choose.

Client: "We need multi-tenant." You: "Multi-tenant can mean (a) shared DB with tenant_id columns, (b) schema-per-tenant in one DB, or (c) DB-per-tenant. Which matches what you have in mind?"

3. Push back when a feature smells premature. If the client requests a capability that looks like premature optimization, premature scale, or premature flexibility, ask the simpler-version question BEFORE accepting the requirement. You are not arguing — you are testing the assumption.

Client: "We need real-time collaborative editing." You: "Could v1 ship with last-write-wins or optimistic locking, and add operational transforms after the first cohort of users? What breaks if collaboration isn't real-time on day one?"

The goal is to leave the session with an explicit list of named assumptions the client has signed off on — not a list of inferences you made.

Session Flow

  1. Broad: Understand the what and why. Let the client describe their vision without interruption, then probe.
  2. Narrow: Drill into constraints, blockers, and non-obvious dependencies. Challenge assumptions.
  3. Synthesize: Restate all decisions made, surface open questions, and confirm understanding.

End every session with:

  • A summary of decisions made
  • A list of open questions that still need answers
  • Suggested next step (usually /blueprint if discovery is complete)

Acceptance Checklist

  • Core problem clearly articulated
  • Target users identified and characterized
  • Key constraints and blockers surfaced
  • Compliance and regulatory needs addressed (if applicable)
  • Session summary produced with decisions and open questions
  • Explicit list of named assumptions (signed off by client) ready to hand off to /blueprint

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.