agentleFS
Sign inStart free
roronoazoroshao369roronoazoroshao369/vibe-coding-os/skills/core/clarify-before-code/SKILL.md

clarify-before-code

Skill1 starsChanged 5 months agoMITView on GitHub

Prevent wasted implementation by resolving ambiguity before edits. Use when requirements, constraints, acceptance criteria, or target files are unclear. User request, observed repo context, possible interpretations. 1. Restate the request in concrete terms. 2. Identify ambiguities that affect implementation. 3. Ask the fewest high-value questions possible — cap at five per round. Skip any question already answered by the request or existing artifacts; prefer precision over breadth. 4. Encode the answers back into the spec artifact. A clarification that lives…

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use when requirements, constraints, acceptance criteria, or target files are unclear.

    line 20
  2. Distinctive:

    Ask the fewest high-value questions possible — cap at five per round.

    line 30
  3. Distinctive:

    If questions are not possible, record conservative assumptions in the spec's Assumptions section.

    line 35
  4. Distinctive:

    Never ask the user to restate what they already said.

    line 44
  5. Distinctive:

    After answers arrive, write them into the spec (not just the reply) and narrow the task statement accordingly.

    line 45
  6. Very specific:

    Use with brainstorming when the request needs option exploration before a spec.

    line 69

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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

Skill: Clarify Before Code

Purpose

Prevent wasted implementation by resolving ambiguity before edits.

When to use

Use when requirements, constraints, acceptance criteria, or target files are unclear.

Inputs

User request, observed repo context, possible interpretations.

Workflow

  1. Restate the request in concrete terms.
  2. Identify ambiguities that affect implementation.
  3. Ask the fewest high-value questions possible — cap at five per round. Skip any question already answered by the request or existing artifacts; prefer precision over breadth.
  4. Encode the answers back into the spec artifact. A clarification that lives only in chat is lost; update the spec's goals, constraints, acceptance criteria, or assumptions so the resolved decision becomes part of the source of truth.
  5. If questions are not possible, record conservative assumptions in the spec's Assumptions section.
  6. Proceed only when the next action is safe.

Question discipline

  • At most five questions per round; ask follow-ups only when ≥2 scenario classes (alternate, exception, recovery, non-functional) remain unclear, with a one-line justification each.
  • Each question must materially change the implementation or the spec; drop the rest.
  • Never ask the user to restate what they already said.
  • After answers arrive, write them into the spec (not just the reply) and narrow the task statement accordingly.

Outputs

Clarifying questions (≤5), the answers encoded back into the spec artifact, or documented conservative assumptions, plus a narrowed task statement.

Failure modes

  • Asking excessive questions.
  • Proceeding despite conflicting requirements.
  • Inventing product behavior.

Verification checklist

  • Ambiguities that affect code are resolved or recorded.
  • No more than five questions were asked per round.
  • Answers are encoded back into the spec artifact, not left only in chat.
  • Assumptions are conservative.
  • The user goal remains unchanged.

Superpowers alignment

Use with brainstorming when the request needs option exploration before a spec.

Related mattpocock-inspired skill: skills/core/grill-user-before-building/SKILL.md for deeper pre-build interviewing.

Spec-driven development layer (github/spec-kit inspiration)

The ≤5-question cap and the "encode answers back into the spec" discipline adapt the clarify idea from github/spec-kit (MIT, GitHub, Inc.) as original wording. No upstream prompt text or CLI is vendored. Related: commands/vibe-spec.md, skills/core/requirements-quality-checklist/SKILL.md.

Ghi chú tiếng Việt

Hỏi tối đa năm câu mỗi vòng, bỏ qua câu đã được trả lời, và quan trọng nhất: ghi câu trả lời NGƯỢC vào artifact spec (goals / constraints / acceptance criteria / assumptions), không để trôi trong chat. Chỉ hỏi khi câu hỏi thực sự thay đổi cách làm. Học ý tưởng từ github/spec-kit (MIT, GitHub, Inc.), viết lại nguyên bản, không copy prompt/CLI.

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.

Check your setup, free