ormastes/simple/arch
Claude Code and agent skill by ormastes/simple
Trust tier 3
- community
Source check pending
Checking that the SKILL.md file is still in the repository.
Install
npx vskill@latest install ormastes/simple/archSKILL.md
The same SKILL.md was published earlier as ormastes/simply/arch.
Arch Skill — Architecture + System Test Design
Prerequisites
| Artifact | Path | If missing | |
|---|---|---|---|
| Requirements | doc/02_requirements/feature/<feature>.md | Run /research first | |
| NFR | doc/02_requirements/nfr/<feature>.md | Run /research first |
Phase 1: Architecture
- Evaluate architecture patterns (ask user which to use)
- Apply MDSOC pattern where appropriate (see
src/compiler/85.mdsoc/) - Output:
doc/04_architecture/<feature>.md
Phase 2: System Test Design
- Generate SPipe BDD tests:
test/03_system/app/<app_name>/feature/<feature>_spec.spl - Follow SPipe rules:
describe,it,expectbuilt-in (no import)- One assertion concept per test
- Clear names:
it "adds two positive numbers":notit "works": """..."""docstrings for generated docs
- Matchers (built-in only):
to_equal,to_be,to_be_nil,to_contain,to_start_with,to_end_with,to_be_greater_than,to_be_less_than - Verify every REQ-NNN has at least one test
- Test plan:
doc/03_plan/sys_test/<feature>.md - For broad lanes, define lower-model sidecars such as Codex Spark, Claude
Haiku, or Claude Sonnet, or mark N/A; record merge owner and final normal/highest-capability reviewer in doc/03_plan/agent_tasks/<feature>.md
- Before sidecars fan out, the best available model defines shared interface
names, manual step("...") flow helper names, setup/checker helper names, and fail-fast placeholders using assert(false) or fail(...)
Quality Check
- Verify SPipe quality (target: A grade) — real assertions, edge cases, full REQ coverage
- Verify generated-manual quality and normal/highest-capability review of
merged sidecar output before handoff
- Ask user: "Should architecture change?"
- If yes, loop back
Outputs
| Artifact | Location | |
|---|---|---|
| Architecture | doc/04_architecture/<feature>.md | |
| System tests | test/03_system/app/<app_name>/feature/<feature>_spec.spl | |
| Test plan | doc/03_plan/sys_test/<feature>.md | |
| Agent tasks | doc/03_plan/agent_tasks/<feature>.md |
Critical Rules
- User must approve architecture before moving to
/design - Every REQ-NNN must have test coverage
- For MCP, LSP, and tool servers, design must cover startup path, hot request path, cache or index strategy, and invalidation
- Treat full-tree scans, repeated file rereads, and per-request subprocesses as first-class design risks
- Define performance budgets and lightweight observability for perf-sensitive features before implementation
Security
- Security scanPassed, 90 of 100Pattern checks on the skill's files.
- Model reviewPassed
Quality evaluation
- Task evaluationNot runNo evaluation yet. Run one in Skill Studio.
Source
- SKILL.md in the repositoryCheckingexamples/05_stdlib/spipe/doc/00_llm_process/skill_command/skills/pipe/design/arch/skill.md
Popularity
- 7-day trend score
- 5
Not refreshed yet.
Works with
Any agent vskill installs to: 56 targets in all. These 39 have a page on this site.
- Claude Code
- Cursor
- GitHub Copilot
- Windsurf
- Codex
- Gemini CLI
- Cline
- Amp
- Roo Code
- Goose
- Continue
- Kilo Code
- Junie
- Trae
- OpenHands
- Qwen Code
- AI Agent
- Tabnine Chat
- Sourcegraph Cody
- Amazon Q
- Firebase Genkit
- Sweep
- Devin
- Phind
- Mentat
- Claude Desktop
- Aider
- GPT Pilot
- Plandex
- bolt.new
- v0
- Replit Agent
- Anse
- TypingMind
- BetterChatGPT
- LibreChat
- Warp AI
- Pieces
- Fig AI
Embed badge
[](https://verified-skill.com/skills/ormastes/simple/arch)