agentleFS
Sign inStart free
felipefontourafelipefontoura/pi-sdd-kit/skills/sdd-idea/SKILL.md

sdd-idea

Skill36 starsChanged 6 months agoMITView on GitHub

Conversational SDD skill for exploring raw product or feature ideas before planning. Use when the user wants to brainstorm, clarify, compare directions, or convert an early idea into a plan or requirements, even without an explicit command. Do not use for implementation or technical task execution.

We split this skill into the rules it gives an agent

23 rules. About 0 are common, 13 are distinctive, and 10 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈13

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 10

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Use this skill to create cognitive clarity before formal planning.

    line 12
  2. Very specific:

    Read ../_shared/references/templates.md only when writing the final artifact.

    line 19
  3. Very specific:

    Use package-level templates/idea.md as the user-facing template when available.

    line 20
  4. Very specific:

    If .ai/strategy/handoff/strategy-brief.md exists, read it as optional upstream context.

    line 22
  5. Very specific:

    Use .ai/sdd/ideas/ for idea artifacts.

    line 25
  6. Very specific:

    Create .ai/sdd/ideas/ if it does not exist.

    line 26
  7. Very specific:

    If .ai/sdd/PLAN.md or .ai/sdd/specs/ exists, briefly inspect them to avoid duplicating an existing direction.

    line 27
  8. Distinctive:

    If using the optional strategy handoff, record it as a source path in the idea artifact.

    line 28
  9. Distinctive:

    Ask one question at a time.

    line 31
  10. Distinctive:

    Prefer multiple-choice questions when useful.

    line 32
  11. Distinctive:

    Do not ask implementation-first questions unless the idea is explicitly technical.

    line 34
  12. Distinctive:

    Ask the user whether to continue exploring, create a PLAN, create REQUIREMENTS directly, or archive the idea.

    line 40
  13. Distinctive:

    Write the idea artifact only after the user confirms the direction.

    line 42
  14. Distinctive:

    Use clear technical product language.

    line 46
  15. Very specific:

    If .ai/sdd/INDEX.md exists, add or update an entry for the idea.

    line 49
  16. Very specific:

    Do not invent statuses; use idea:captured for saved ideas.

    line 50
  17. Very specific:

    create .ai/sdd/PLAN.md for medium projects

    line 56
  18. Very specific:

    create .ai/sdd/specs/NNN-feature/requirements.md directly for small features

    line 57
  19. Distinctive:

    Do not create requirements, design, tasks, or code inside this skill unless the user explicitly changes the requested phase.

    line 61
  20. Distinctive:

    Do not collapse exploration into a plan too early.

    line 62
  21. Distinctive:

    Do not treat IDEA as approved scope.

    line 63
  22. Distinctive:

    Do not proceed without explicit user confirmation before writing the final idea file.

    line 64
  23. Distinctive:

    Do not silently promote or infer feature-spec approval states from IDEA artifacts.

    line 65

≈ “Common” is an estimate: at least 1 in 5 of the 308,232 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

SDD Idea

Explore an idea before turning it into a plan, requirements, or code.

Purpose

Use this skill to create cognitive clarity before formal planning. IDEA is divergent: it explores the problem space, users, alternatives, value signals, constraints, risks, and possible directions. It does not commit the user to a roadmap or implementation.

Workflow

  1. Load the SDD reference.
    • Read ../_shared/references/sdd-practical.md.
    • Apply the Language Policy: respond and write artifacts in the user's initial chat language while keeping skill instructions and templates in EN-US.
    • Read ../_shared/references/templates.md only when writing the final artifact.
    • Use package-level templates/idea.md as the user-facing template when available.
    • Load relevant .ai/steering/*.md files when present, especially product, principles, or domain context.
    • If .ai/strategy/handoff/strategy-brief.md exists, read it as optional upstream context; use it to seed exploration, but do not require it or treat it as approved SDD scope.
  2. Establish the workspace.
    • Use .ai/sdd/ideas/ for idea artifacts.
    • Create .ai/sdd/ideas/ if it does not exist.
    • If .ai/sdd/PLAN.md or .ai/sdd/specs/ exists, briefly inspect them to avoid duplicating an existing direction.
    • If using the optional strategy handoff, record it as a source path in the idea artifact.
  3. Explore the raw idea.
    • Ask one question at a time.
    • Prefer multiple-choice questions when useful.
    • Cover at least: problem, target users, current alternatives, desired outcome, constraints, success signals, and risk.
    • Do not ask implementation-first questions unless the idea is explicitly technical.
  4. Present directions.
    • Propose 2-3 possible directions.
    • Include trade-offs: speed, complexity, user value, and risk.
    • Recommend one direction, but preserve dissenting alternatives.
    • Ask the user whether to continue exploring, create a PLAN, create REQUIREMENTS directly, or archive the idea.
  5. Write the idea artifact only after the user confirms the direction.
    • Determine the next numeric prefix in .ai/sdd/ideas/.
    • Use NNN-kebab-name.md.
    • Fill the Idea Template from ../_shared/references/templates.md.
    • Use clear technical product language; avoid implementation commitments.
  6. Update optional index.
    • If .ai/sdd/INDEX.md exists, add or update an entry for the idea.
    • Do not invent statuses; use idea:captured for saved ideas.

Output

  • .ai/sdd/ideas/NNN-name.md
  • Optional recommendation for next command:
    • create .ai/sdd/PLAN.md for medium projects
    • create .ai/sdd/specs/NNN-feature/requirements.md directly for small features

Critical Rules

  • Do not create requirements, design, tasks, or code inside this skill unless the user explicitly changes the requested phase.
  • Do not collapse exploration into a plan too early.
  • Do not treat IDEA as approved scope; it is a thinking artifact.
  • Do not proceed without explicit user confirmation before writing the final idea file.
  • Do not silently promote or infer feature-spec approval states from IDEA artifacts.

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