Skip to content

getting-started.md

Latest commit

 

History

History
133 lines (92 loc) · 4.75 KB

getting-started.md

File metadata and controls

133 lines (92 loc) · 4.75 KB

Getting Started

Get productive in 15 minutes.

Prerequisites

  • Claude Code CLI — install guide
  • Git — to clone the repository
  • Python 3.10+ and uv — for validation scripts (uv sync)

Install the Plugin

# 1. Clone
git clone https://github.com/Modular-Earth-LLC/solutions-architecture-agent.git
cd solutions-architecture-agent

# 2. Install Python test dependencies
uv sync

# 3. Load as Claude Code plugin
claude --plugin-dir .

# Or install persistently (see docs/local-setup.md)
# claude plugin marketplace add .
# claude plugin install solutions-architecture-agent@solutions-architecture-agent

Claude automatically loads CLAUDE.md and .claude/rules/ configuration. Skills appear as solutions-architecture-agent:<skill-name>.

Run Your First Skill

Option A — Slash command:

/requirements

Option B — Natural language:

I need to design a healthcare data platform that modernizes our legacy systems.

The agent classifies your intent and routes to the appropriate skill. It asks structured questions, you answer, and it produces validated JSON output in knowledge_base/. After each skill completes, it presents a human checkpoint: summary, deliverables, and suggested next skill.

Want to see what a full engagement produces? Read the sample proposal or explore the complete example outputs.

Understanding the System

Single agent with skills. Each skill handles one phase of the SA lifecycle:

  1. /requirements — Discover what the client needs
  2. /architecture — Design the system
  3. /data-model — Model the data layer
  4. /security-review — STRIDE threat modeling + compliance
  5. /integration-plan — APIs, migration, legacy bridging
  6. /estimate — LOE, cost, team composition
  7. /project-plan — Phased roadmap with sprints
  8. /proposal — Assemble SOW from all upstream deliverables
  9. /review — LLM-as-judge quality review

Skills communicate through JSON files in knowledge_base/ — each skill reads from upstream files and writes to its own file.

Engagement Flows

Choose the flow that matches your engagement type:

Flow Skills When
Greenfield req → arch → dm → sr → est → ppl → pro → rv New system from scratch
Migration req → ip → arch → dm → sr → est → ppl → pro → rv Legacy modernization
Streamlined req → arch → est → pro Small/time-constrained
Assessment req → arch → [sr] → pro Discovery-only
Quick Qualify req (quick tier) Pipeline qualification
Direct Delivery single skill (QUICK) Single-document tasks, interview prep
Rapid Assessment req (QUICK) → arch (QUICK) → pro (QUICK) Same-day turnaround
Custom Document selective skills (QUICK) → proposal --type custom User-specified format and sections

You don't need to run every skill — the agent checks prerequisites and tells you what's missing.

Depth Tiers

Tier When to Use KB Files? Sub-Agents? Typical Output
QUICK Time-constrained, interview prep, same-day No No 1-4 page doc
STANDARD Client engagements, complete SA work Yes Conditional Full KB + output
COMPREHENSIVE High-stakes, enterprise procurement Yes Yes Full KB + review

Use --depth QUICK flag to set tier explicitly. Otherwise, the agent determines tier from your answers during scope negotiation.

Validation

Run the validation scripts to check file integrity. See tests/README.md for details and expected output.

python tests/validate_knowledge_base.py
python tests/validate_consistency.py
python tests/test_plugin_structure.py
python tests/test_engagement_flow.py
python tests/test_skill_independence.py

Security & Sensitive Data

  • Use the private/ directory for any sensitive files (gitignored)
  • See private/README.md for security guidelines
  • Never commit API keys, credentials, or PII to version control
  • Verify with git status before committing

Next Steps