agentleFS
michaelalbermichaelalber/ai-toolkit/skills/qraspi-questions/SKILL.md

qraspi-questions

Skill3 starsChanged 4 months agoMITView on GitHub

QRASPI Questions phase -- surface what is unknown about a NEW system before any research, architecture, or skeleton begins. Use for "/qraspi-questions <project>", "new project from scratch", "greenfield X", "questions for a brand-new system". Do NOT use for QRSPI (an EXISTING codebase / adding a feature) -- that routes to qrspi-questions. Do NOT use for the deprecated RPI workflow.

The file

We split this skill into the rules it gives an agent

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

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    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:

    ASK, never answer -- this phase produces questions, not solutions, research, or architecture

    line 29
  2. Very specific:

    STOP after writing questions.md -- the human answers inline before Research begins

    line 33

≈ “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

QRASPI Questions

"If you do not know how to ask the right question, you discover nothing." -- W. Edwards Deming

Core Philosophy

The Questions phase is the first alignment gate of QRASPI's greenfield (V0/V1) path. Before any domain research, architecture, or skeleton, the agent surfaces every unknown about the new system as a targeted question and STOPS for the human to answer. Greenfield has no codebase to constrain the answers, so the danger is a narrow question set that silently commits the design to one shape. This phase defends against that with a fixed six-category checklist: it enumerates a question for every greenfield concern whether or not the user named it. A skipped category becomes an unexamined assumption that cascades into the ADRs; surfacing it here costs one edit in questions.md, not an architecture rewrite later.

Non-Negotiable Constraints:

  1. ASK, never answer -- this phase produces questions, not solutions, research, or architecture
  2. Cover ALL SIX greenfield categories -- functional scope · quality attributes (the -ilities) · integration / external systems · compliance / regulatory · deployment / runtime target · data & domain model -- enumerate a question for every category, named by the user or not
  3. STOP after writing questions.md -- the human answers inline before Research begins
  4. One project folder per project -- every QRASPI artifact co-locates there
  5. CONTEXT BUDGET: keep utilization under 40%. At 60%, write questions.md with progress and tell the user to start a fresh session.

Workflow

PRE-FLIGHT
    [ ] Determine the project slug (kebab-case) and today's date
    [ ] Project folder = thoughts/shared/qraspi/YYYY-MM-DD-{slug}/   (create if absent)
    [ ] If an ANSWERED questions.md already exists here, this phase is DONE -> route to /qraspi-research

SURFACE
    Walk the SIX greenfield categories in order; for EACH, write at least one specific,
    answerable question -- never a vague prompt, never skip a category:
      functional scope · quality attributes (-ilities) · integration / external systems ·
      compliance / regulatory · deployment / runtime target · data & domain model
    Mark any question that BLOCKS architecture as [BLOCKING]

WRITE
    Create thoughts/shared/qraspi/YYYY-MM-DD-{slug}/questions.md
    Use references/questions-template.md
    Set status: awaiting-answers

STOP
    Tell the user: answer the questions inline in questions.md, then start a NEW session
    and run /qraspi-research. Do NOT proceed to Research yourself.

Exit criteria: questions.md written with status awaiting-answers; all six greenfield categories carry at least one question with blocking items flagged; user told to answer inline and start a fresh Research session.

State Block

<qraspi-questions-state>
phase: PRE-FLIGHT | SURFACE | WRITE | STOP | COMPLETE
project_slug: [kebab-slug]
project_folder: thoughts/shared/qraspi/YYYY-MM-DD-{slug}/
areas_covered: [functional | quality-attributes | integration | compliance | deployment | domain]
                # all six MUST be present before WRITE
question_count: [count]
blocking_count: [count]
context_budget: under-40 | approaching-60 | checkpoint-now
status: awaiting-answers | complete
</qraspi-questions-state>

Output Template

See references/questions-template.md for the full questions.md structure and frontmatter.

Integration with Other Skills

SkillRelationship
qraspi-researchNext phase. Consumes the ANSWERED questions.md to scope the landscape map.
qraspi-architectureDownstream. Every ADR's decision traces back to an answer captured here.
qrspi-questionsBrownfield sibling. Use it instead when the system already EXISTS and you are adding a feature.
spec-coachUse instead when you want an interactive design conversation, not a one-shot question dump.

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.