agentleFS
aman-bhandariaman-bhandari/claude-code-agent-skills-framework/.claude/skills/architect-first/SKILL.md

architect-first

Skill0 starsChanged 7 months agoMITView on GitHub

Forces a planning phase before any code is written. Triggers on /architect or at the start of BUILD sessions. Requires the student to articulate inputs, outputs, failure modes, and data flow before implementation begins. Inspired by Copilot Plan Mode and production engineering practices.

4 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 5 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 5

    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:

    No code gets written until the architecture is clear.

    line 8
  2. Distinctive:

    Gate: Only after the student has a clear, challenged plan → "Go build it."

    line 34
  3. Distinctive:

    Do NOT write pseudocode, code, or architecture diagrams.

    line 39
  4. Distinctive:

    Ask "why" for every design decision.

    line 41
  5. Distinctive:

    If the student skips this and jumps to code, stop them: "Architecture first."

    line 43

“Common” is measured: kept by at least 6 in 10 of the 4 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

Architect First -- Plan Before Code

No code gets written until the architecture is clear. If you can't explain the data flow, you're not ready to code.

Trigger

  • /architect or /architect <exercise>
  • Automatically before BUILD: sessions

Execution

  1. Read the exercise to understand what's being built.
  2. The student articulates the plan (not Claude):
    • What are the inputs? (types, formats, edge cases)
    • What are the outputs? (exact format, error states)
    • What are the failure modes? (what can go wrong?)
    • What's the data flow? (step by step, plain English)
    • What standard library modules will you use and why?
  3. Challenge the plan:
    • "What happens when the input is empty?"
    • "What happens at 10x scale?"
    • "Where would you add logging?"
    • "How would you test this?"
    • "In a production data pipeline, how would this work?"
    • "How will you structure your functions so each can be tested independently?"
  4. Gate: Only after the student has a clear, challenged plan → "Go build it."

Rules

  • This skill is a gate, not a guide. The student articulates the plan.
  • Do NOT write pseudocode, code, or architecture diagrams.
  • The plan is in natural language only.
  • Ask "why" for every design decision.
  • The plan doesn't need to be perfect. It needs to exist.
  • If the student skips this and jumps to code, stop them: "Architecture first."

Production Patterns Reference

See PATTERNS.md for common architecture patterns to ask about during planning.

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.