agentleFS
caiaffacaiaffa/claude-code-ultimate-engineering-system/skills/architecture-decisions/SKILL.md

architecture-decisions

Skill17 starsChanged 58 days agoMITView on GitHub

Evaluate technical options with explicit trade-offs across reliability, delivery speed, operability, cost, security, and long-term maintainability.

6 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

1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 4 edited versions we read keep these.

  • 1

    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:

    Only one option seriously considered.

    line 36

“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

Mission

Produce architecture recommendations grounded in constraints, trade-offs, and system consequences rather than technical preference.

When to use

  • Choosing between implementation approaches.
  • Deciding service boundaries.
  • Evaluating platform or infrastructure changes.
  • Comparing sync vs async patterns.
  • Writing ADRs or design decisions.

Handoff

  • Receives from: CLAUDE.md orchestrator (architecture category).
  • Hands off to: adr-challenger (for challenge), then backend-platform-engineer (for implementation).

Before answering

Identify: business urgency, scale/traffic profile, team maturity, operational burden tolerance, budget sensitivity, compatibility requirements.

Decision framework

For each option, evaluate on these axes (1-5 scale):

  • Complexity to build — how hard to implement correctly?
  • Complexity to operate — how hard to run, debug, and maintain?
  • Reliability — what happens when things go wrong?
  • Performance — does it meet latency/throughput needs?
  • Cost — build cost + run cost + maintenance cost?
  • Reversibility — how hard to undo if wrong?
  • Security — does it increase or decrease attack surface?
  • Team fit — can this team realistically build and operate this?

Red flags

  • Solution chosen before problem is clearly stated.
  • Only one option seriously considered.
  • "We can optimize later" for a known bottleneck.
  • Architecture requires discipline the team hasn't demonstrated.
  • Elegant design with no operational story.

Output format

  1. Context (what is happening)
  2. Problem (what specifically needs deciding)
  3. Constraints (hard limits)
  4. Options considered (minimum 2 real options)
  5. Trade-off matrix (axes above, scored)
  6. Recommendation with justification
  7. Risks with severity and mitigation
  8. Follow-up actions with owner

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.