{ "markdown": "

\n
\n \n \n \n \"Architect\n \n
\n

\n\n

\n Ace the Claude Certified Architect exam\n

\n\n

\n Adaptive certification prep powered by the Model Context Protocol.
\n 390 questions. Guided capstone build. 30 concept handouts. 6 reference projects. Practice exams. Interactive dashboard. Spaced repetition. Zero sycophancy.\n

\n\n

\n \"npm\n \"npm\n \"GitHub\n \"License\"\n \"Glama\n

\n\n

\n \n \"Connectry\n \n

\n\n

\n Quick Start •\n Learning Path •\n Interactive Experience •\n Assessment •\n Handouts •\n Practice •\n Capstone Build •\n Practice Exams •\n Dashboard •\n Reference Projects •\n Tools •\n Architecture\n

\n\n---\n\n## What is Architect Cert?\n\nArchitect Cert is a free, open-source [MCP](https://modelcontextprotocol.io/) server that turns Claude into your personal certification tutor for the **Claude Certified Architect — Foundations** exam. No courses, no slides, no video lectures — just ask Claude and study.\n\nIt ships with:\n- **390 scenario-based questions** across all 5 exam domains and 30 task statements\n- **Interactive clickable UI** — answer questions with A/B/C/D buttons, select follow-ups, pick domains with checkboxes — all inside Claude\n- **Lessons-first assessment** — learn each domain's concepts before being tested on them\n- **Guided capstone build** — shape your own project, then build it step-by-step while learning every task statement hands-on\n- **30 concept handouts** — one per task statement, with code examples and common mistakes\n- **6 reference projects** — runnable TypeScript codebases demonstrating each domain in practice\n- **Practice exams** — 60-question weighted exams with history tracking and improvement trends\n- **Progress dashboard** — glassmorphism visual dashboard with mastery levels, exam history chart, activity timeline\n- **Visual progress tracking** — todo checklists track your progress through assessments, exams, and capstone builds in real time\n- **Interactive follow-ups** — wrong answer? Click to see code examples, concept lessons, handouts, or reference projects\n- **PDF generation** — branded handout PDFs with the Architect Cert logo for offline study\n- **Spaced repetition** — SM-2 algorithm schedules reviews at optimal intervals\n- **Deterministic grading** — pure function grading, no LLM judgment, zero sycophancy\n\nEverything runs locally. No cloud, no accounts, no telemetry.\n\n
\n\n## Quick Start\n\n### 1. Install\n\n```bash\nnpm install -g connectry-architect-mcp\n```\n\n### 2. Configure Your MCP Client\n\n
\nClaude Code — One-liner\n\n```bash\nclaude mcp add connectry-architect -- connectry-architect-mcp\n```\n\nThat's it. Restart Claude Code and the server starts automatically.\n\n
\n\n
\nClaude Code — Manual config\n\nAdd to `.mcp.json` in your project or `~/.claude.json` globally:\n\n```json\n{\n \"mcpServers\": {\n \"connectry-architect\": {\n \"command\": \"connectry-architect-mcp\"\n }\n }\n}\n```\n\nRestart Claude Code. The server starts automatically when Claude loads.\n\n
\n\n
\nClaude Desktop — macOS\n\nAdd to `~/Library/Application Support/Claude/claude_desktop_config.json`:\n\n```json\n{\n \"mcpServers\": {\n \"connectry-architect\": {\n \"command\": \"connectry-architect-mcp\"\n }\n }\n}\n```\n\n> **Tip:** If you use `nvm`, you may need to specify the full path to the binary:\n> ```json\n> \"command\": \"/Users/yourname/.nvm/versions/node/v22.20.0/bin/connectry-architect-mcp\"\n> ```\n\nRestart Claude Desktop. You'll see the MCP tools icon appear in the chat input.\n\n
\n\n
\nClaude Desktop — Windows\n\nAdd to `%APPDATA%\\Claude\\claude_desktop_config.json`:\n\n```json\n{\n \"mcpServers\": {\n \"connectry-architect\": {\n \"command\": \"connectry-architect-mcp\"\n }\n }\n}\n```\n\nRestart Claude Desktop. You'll see the MCP tools icon appear in the chat input.\n\n
\n\n
\nAny MCP-compatible client\n\nArchitect Cert works with any client that supports the [Model Context Protocol](https://modelcontextprotocol.io/). Configure it as a stdio server:\n\n- **Command:** `connectry-architect-mcp`\n- **Arguments:** none\n- **Transport:** stdio\n\nThe server exposes 18 tools, 8 prompts, and 3 resource types.\n\n
\n\n### 3. Start Studying\n\nRestart your MCP client and start chatting:\n\n| What you want | What to ask Claude |\n|---------------|-------------------|\n| Start from scratch | *\"Start an assessment to figure out where I stand\"* |\n| Practice questions | *\"Give me a practice question\"* |\n| Focus on a domain | *\"Give me a question about agentic architecture\"* |\n| Learn a concept first | *\"Teach me about task 2.3 — tool provisioning\"* |\n| Build your own capstone | *\"I want to start a guided capstone build\"* |\n| Take a practice exam | *\"I want to take a practice exam\"* |\n| Check your progress | *\"Show my study progress\"* |\n| Show the dashboard | *\"Show my dashboard\"* |\n| Get a study plan | *\"What should I study next?\"* |\n| Explore a reference project | *\"Show me a reference project for domain 1\"* |\n| Generate PDF handouts | Run `npm run generate:pdfs` in the project directory |\n| Reset and start over | *\"Reset my progress\"* |\n\n
\n\n## Learning Path\n\nArchitect Cert is designed to follow a natural progression. Here's the recommended order:\n\n```\n1. Assessment → Baseline your knowledge across all 5 domains\n ↓\n2. Learn concepts → Read handouts for weak domains before practicing\n ↓\n3. Adaptive practice → Questions prioritized by weakness, reviews, new material\n ↓\n4. Follow-ups → Dive into code examples, concept lessons, reference projects\n ↓\n5. Capstone build → Build your own project covering all 30 task statements\n ↓\n6. Practice exams → Simulate the real exam (60 questions, scored out of 1000)\n ↓\n7. Review & repeat → Dashboard tracks mastery, spaced repetition handles scheduling\n```\n\nYou can jump to any stage at any time — the system adapts. But the path above gives you the most structured experience.\n\n
\n\n## Interactive Experience\n\nArchitect Cert uses Claude's built-in interactive UI to make studying feel natural. No typing answer letters — just click.\n\n### Clickable Answer Selection\n\nEvery question presents **clickable A/B/C/D buttons** directly in Claude. You tap your answer instead of typing it. If the question includes code in the scenario, a **code preview** appears alongside each option so you can reference it while deciding.\n\n### Follow-Up Actions\n\nAfter every answer, clickable follow-up buttons appear:\n\n**After a wrong answer:**\n- Got it, next question\n- Explain with a code example\n- Show me the concept lesson\n- Show me the handout\n- Show me in the reference project\n\n**After a correct answer:**\n- Next question\n- Explain why the others are wrong\n\n### Multi-Select Domain Picker\n\nWhen requesting a study plan, you can select multiple domains to focus on using **checkboxes** — no need to list them out by name.\n\n### Visual Progress Checklists\n\nLong flows like the assessment (15 questions), practice exams (60 questions), and capstone build (18 steps) create a **visual todo checklist** that updates in real time as you progress. You always know where you are and what's left.\n\n### Skip & Other\n\nEvery question selection includes a **Skip** button to move on and an **Other** option to ask a free-form question. If you use \"Other\" to ask something, Claude answers your question and then **re-presents the same quiz question** — you never lose your place. \"Skip\" moves to the next question; skipped questions remain unanswered and will reappear later.\n\n
\n\n## Initial Assessment\n\nStart with a **15-question diagnostic** (3 per domain) that determines your learning path. The assessment works differently from raw practice:\n\n### Lessons-First Flow\n\nWhen the assessment reaches a **new domain** for the first time, it pauses to show you the **concept handout** before asking questions. This means you learn the key ideas, see code examples, and understand common mistakes *before* being tested — no guessing blindly.\n\nThe flow for each domain:\n1. Concept handout is shown (one-time per domain)\n2. 3 questions presented one at a time (easy, medium, hard)\n3. Each answer graded immediately with explanation\n4. Move to the next domain\n\n### Path Assignment\n\nBased on your overall accuracy:\n\n| Score | Path | Description |\n|-------|------|-------------|\n| **< 60%** | Beginner-Friendly | Starts with fundamentals, builds up gradually. Focuses on easy and medium questions first. |\n| **>= 60%** | Exam-Weighted | Focuses on high-weight domains first (D1 at 27%, D3 & D4 at 20% each). Targets weak areas aggressively. |\n\n### Progress Tracking\n\nThe assessment creates a **15-item visual checklist** (Q1 through Q15, grouped by domain) that updates after each answer. You always see exactly how far you've progressed.\n\n
\n\n## Concept Handouts\n\nEvery task statement has a **concept handout** — a structured study document (~500-800 words) that covers:\n\n- **Concept** — The core idea, mental model, and when/why to use it\n- **Code Example** — Realistic TypeScript demonstrating the pattern\n- **Common Mistakes** — The 3-5 most frequent errors (which map to exam wrong answers)\n- **References** — Links to Anthropic's official documentation\n\n### Reading Handouts in Claude\n\n```\nYou: \"Show me the handout for task 2.3\"\n\nClaude: [loads the full handout with concept, code, mistakes, references]\n```\n\n### When Handouts Appear Automatically\n\n- **During assessment** — shown before the first question of each new domain\n- **After a wrong answer** — \"Show me the handout\" is one of the follow-up options\n- **During capstone build** — relevant handouts are available at each build step\n\n### Generating PDF Handouts\n\nGenerate branded PDFs for all 30 handouts for offline study:\n\n```bash\ncd connectrylab-architect-cert-mcp\nnpm run generate:pdfs\n```\n\nThis creates 30 PDFs in `generated/handouts/` with:\n- Architect Cert logo and domain label in the header\n- Clean formatting with syntax-highlighted code blocks\n- \"Connectry LABS — Claude Certified Architect Exam Prep — Free & Open Source\" footer\n\n
\n\n## Adaptive Practice\n\nEvery practice question is selected by a **three-priority algorithm**:\n\n1. **Overdue reviews** — Spaced repetition items due for review today\n2. **Weak areas** — Topics where your mastery is below 50%\n3. **New material** — Fresh questions from your recommended domain\n\n### Filtering\n\nYou can filter practice questions by domain and/or difficulty:\n\n```\nYou: \"Give me a hard question about prompt engineering\"\n\nClaude: [presents a hard Domain 4 question with clickable A/B/C/D buttons]\n```\n\n### Interactive Follow-Ups\n\nAfter every answer, you get clickable follow-up options. Each option dives deeper into the concept — then brings you right back to your quiz. You never leave your study flow to look something up.\n\n### Mastery Levels\n\nEach of the 30 task statements has an independent mastery level:\n\n| Level | Criteria | What it means |\n|-------|----------|---------------|\n| **Unassessed** | No attempts yet | You haven't seen questions on this topic |\n| **Weak** | < 50% accuracy | Needs significant study — questions resurface frequently |\n| **Developing** | 50-69% accuracy | Making progress — keep practicing |\n| **Strong** | 70-89% accuracy | Good understanding — review intervals are longer |\n| **Mastered** | >= 90% accuracy, 5+ attempts, 3+ consecutive correct | Exam-ready — rare reviews |\n\n### Spaced Repetition (SM-2)\n\nThe [SM-2 algorithm](https://en.wikipedia.org/wiki/SuperMemo#Description_of_SM-2_algorithm) schedules review intervals:\n\n- **First review:** 1 day after answering\n- **Second review:** 3 days after first review\n- **Subsequent reviews:** Previous interval x ease factor (starts at 2.5)\n- **Wrong answer:** Interval resets, ease factor decreases by 0.2 (floor: 1.3)\n- **Correct answer:** Ease factor increases by 0.1\n\nDifficult questions come back often. Easy ones space out to weeks or months.\n\n
\n\n## Guided Capstone Build\n\nThe most hands-on way to learn — build your own project from scratch while covering all 30 task statements. Instead of just answering questions, you architect a real system themed to your own idea.\n\n### How It Works\n\nThe capstone build has three phases:\n\n**Phase 1 — Project Shaping**\n\nYou describe a project idea (e.g., \"a multi-agent code review system\"). Claude analyzes your idea against all 30 architectural criteria and identifies gaps. You refine together until every task statement is covered.\n\n```\nYou: \"I want to start a guided capstone build\"\n\nClaude: [presents the 30 criteria across all 5 domains]\n Describe your project idea and I'll analyze coverage.\n\nYou: \"A multi-agent code review system that analyzes PRs\"\n\nClaude: Your idea naturally covers 24/30 criteria. To cover the\n remaining 6, I'd suggest adding: [specific suggestions\n mapped to task statements]\n```\n\n**Phase 2 — Interleaved Build (18 steps)**\n\nEach step follows the same pattern:\n\n1. **Quiz** — 2-3 questions on the task statements you're about to build (clickable A/B/C/D)\n2. **Build** — Claude generates the file's code, themed to your project\n3. **Walkthrough** — Line-by-line explanation mapping code to task statements\n\nA **visual 18-step checklist** tracks your progress in real time.\n\nThe 18 steps build incrementally:\n\n| Steps | What you build | Task Statements |\n|-------|---------------|-----------------|\n| 1-2 | Project config (CLAUDE.md, package.json) | 3.1-3.4 |\n| 3-5 | MCP server, tools, error handling | 2.1-2.5 |\n| 6-10 | Agentic loop, subagents, hooks, workflows, sessions | 1.1-1.7 |\n| 11-13 | Prompts: system, extraction, batch processing | 4.1-4.6 |\n| 14-18 | Context: preservation, triggers, propagation, scratchpad, confidence | 5.1-5.6 |\n\nEvery quiz answer feeds into the same spaced repetition and mastery tracking as regular practice.\n\n**Phase 3 — Final Review**\n\nAfter step 18, you get a complete coverage map: all 30 task statements, where each is demonstrated in your project, and your quiz performance per domain. Weak areas are flagged for further study.\n\n### Capstone Build Tools\n\n| Tool | What it does |\n|------|-------------|\n| `start_capstone_build` | See the 30 criteria, describe your theme, refine until coverage is complete |\n| `capstone_build_step` | Drive the build: confirm, quiz, build, next, status, or abandon |\n| `capstone_build_status` | Check your progress — current step, criteria coverage, quiz performance |\n\n### How It Connects to Everything Else\n\n- Quiz answers during the build use the same `submit_answer` grading and SM-2 scheduling\n- After any quiz question, you can use the same follow-up options (code example, concept lesson, handout, reference project)\n- The reference projects show how the capstone structure looks when complete\n- Progress persists across sessions — pick up where you left off\n\n
\n\n## Practice Exams\n\nFull 60-question exams that simulate the real certification:\n\n| Detail | Value |\n|--------|-------|\n| Total questions | 60 |\n| D1: Agentic Architecture | 16 questions (27%) |\n| D2: Tool Design & MCP | 11 questions (18%) |\n| D3: Claude Code Config | 12 questions (20%) |\n| D4: Prompt Engineering | 12 questions (20%) |\n| D5: Context & Reliability | 9 questions (15%) |\n| Scoring | 0-1000, passing at 720 |\n| Question selection | Fresh set each time — avoids repeating your most recent attempt |\n| UI | Clickable A/B/C/D buttons with code previews |\n| Progress | Visual 60-item checklist updated after each answer |\n\nAll attempts are saved with per-domain score breakdowns and improvement trends.\n\n
\n\n## Progress Dashboard\n\nArchitect Cert includes a **glassmorphism visual dashboard** that renders directly inside Claude via Claude Preview.\n\n### What It Shows\n\n- **Readiness Ring** — Overall exam readiness percentage in a circular progress indicator\n- **Domain Mastery Grid** — 5 cards with progress bars for each exam domain\n- **Exam History Chart** — Line chart plotting your practice exam scores over time, with the 720 passing score marked\n- **Recent Activity Timeline** — Your last 10 answers with correct/incorrect indicators\n- **Capstone Progress** — Current build step and completion percentage\n- **Quick Action Buttons** — Jump to practice, exam, study plan, or capstone\n\n### How to Open\n\n```\nYou: \"Show my dashboard\"\n\nClaude: [opens the branded dashboard in Claude Preview]\n```\n\nThe dashboard can be reopened at any time by asking Claude again. It always reflects your latest data.\n\n### Text Fallback\n\nFor MCP clients that don't support Claude Preview, the dashboard tool also returns a **text summary** with mastery percentages and exam stats.\n\n
\n\n## Reference Projects\n\nArchitect Cert includes **6 complete reference projects** — runnable TypeScript codebases that demonstrate certification concepts in real code. Every file has a header comment mapping it to specific task statements.\n\n| Project | Focus | Files | What You'll See |\n|---------|-------|-------|-----------------|\n| **Capstone** | All 5 domains | 24 | Full multi-agent support system with MCP server, coordinator, subagents, prompt engineering, context management, and hooks |\n| **D1 — Agentic Loop** | Domain 1 | 10 | Multi-agent research coordinator with agentic loops, subagent spawning, hooks, session management, and task decomposition |\n| **D2 — Tool Design** | Domain 2 | 12 | MCP server with split tools, structured errors, agent-scoped tool distribution, resources, and built-in tool patterns |\n| **D3 — Claude Code Config** | Domain 3 | 14 | Complete config reference: CLAUDE.md hierarchy, slash commands, path rules, CI/CD workflows — not runnable code, but a real config layout |\n| **D4 — Prompt Engineering** | Domain 4 | 11 | Data extraction pipeline with explicit criteria, few-shot, structured output, validation-retry, batch processing, and multi-pass review |\n| **D5 — Context Manager** | Domain 5 | 14 | Long-session patterns: context preservation, scratchpad, subagent delegation, escalation, error propagation, confidence calibration, provenance |\n\n### How to Access\n\n```\nYou: \"Show me a reference project for domain 1\"\n\nClaude: [returns the project README, file listing, and architecture walkthrough]\n```\n\n### How They Connect to the Study Flow\n\nWhen you get a question wrong, one of the follow-up options is **\"Show me in the reference project\"** — this takes you straight to the relevant domain project so you can see the concept implemented in real code. Then you jump back to your quiz.\n\n
\n\n## Study Plan\n\nGet personalized study recommendations based on your performance, exam weights, and spaced repetition schedule.\n\n### Multi-Select Domain Focus\n\nWhen requesting a study plan, you can pick specific domains to focus on using **checkboxes**:\n\n```\nYou: \"What should I study next?\"\n\nClaude: [shows domain checkboxes — select the ones you want to focus on]\n [generates a personalized study plan with a visual checklist]\n```\n\nThe study plan creates a **visual checklist** so you can track your progress through each recommendation.\n\n
\n\n## Exam Domains\n\nThe Claude Certified Architect — Foundations exam covers 5 domains:\n\n| # | Domain | Weight | Tasks | Questions |\n|---|--------|--------|-------|-----------|\n| 1 | Agentic Architecture & Orchestration | 27% | 7 | 91 |\n| 2 | Tool Design & MCP Integration | 18% | 5 | 65 |\n| 3 | Claude Code Configuration & Workflows | 20% | 6 | 78 |\n| 4 | Prompt Engineering & Structured Output | 20% | 6 | 78 |\n| 5 | Context Management & Reliability | 15% | 6 | 78 |\n| | **Total** | **100%** | **30** | **390** |\n\n### 30 Task Statements\n\n
\nDomain 1 — Agentic Architecture & Orchestration (7 tasks, 91 questions)\n\n| Task | Description |\n|------|-------------|\n| 1.1 | Design and implement agentic loops for autonomous task execution |\n| 1.2 | Orchestrate multi-agent systems with coordinator-subagent patterns |\n| 1.3 | Configure subagent invocation, context passing, and spawning |\n| 1.4 | Implement multi-step workflows with enforcement and handoff patterns |\n| 1.5 | Apply Agent SDK hooks for tool call interception and data normalization |\n| 1.6 | Design task decomposition strategies for complex workflows |\n| 1.7 | Manage session state, resumption, and forking |\n\n
\n\n
\nDomain 2 — Tool Design & MCP Integration (5 tasks, 65 questions)\n\n| Task | Description |\n|------|-------------|\n| 2.1 | Design effective tool interfaces with clear descriptions and boundaries |\n| 2.2 | Implement structured error responses for MCP tools |\n| 2.3 | Distribute tools appropriately across agents and configure tool choice |\n| 2.4 | Integrate MCP servers into Claude Code and agent workflows |\n| 2.5 | Select and apply built-in tools effectively |\n\n
\n\n
\nDomain 3 — Claude Code Configuration & Workflows (6 tasks, 78 questions)\n\n| Task | Description |\n|------|-------------|\n| 3.1 | Configure CLAUDE.md files with appropriate hierarchy and scoping |\n| 3.2 | Create and configure custom slash commands and skills |\n| 3.3 | Apply path-specific rules for conditional convention loading |\n| 3.4 | Determine when to use plan mode vs direct execution |\n| 3.5 | Apply iterative refinement techniques for progressive improvement |\n| 3.6 | Integrate Claude Code into CI/CD pipelines |\n\n
\n\n
\nDomain 4 — Prompt Engineering & Structured Output (6 tasks, 78 questions)\n\n| Task | Description |\n|------|-------------|\n| 4.1 | Design prompts with explicit criteria to improve precision |\n| 4.2 | Apply few-shot prompting to improve output consistency |\n| 4.3 | Enforce structured output using tool use and JSON schemas |\n| 4.4 | Implement validation, retry, and feedback loops |\n| 4.5 | Design efficient batch processing strategies |\n| 4.6 | Design multi-instance and multi-pass review architectures |\n\n
\n\n
\nDomain 5 — Context Management & Reliability (6 tasks, 78 questions)\n\n| Task | Description |\n|------|-------------|\n| 5.1 | Manage conversation context to preserve critical information |\n| 5.2 | Design effective escalation and ambiguity resolution patterns |\n| 5.3 | Implement error propagation strategies across multi-agent systems |\n| 5.4 | Manage context effectively in large codebase exploration |\n| 5.5 | Design human review workflows and confidence calibration |\n| 5.6 | Preserve information provenance and handle uncertainty in synthesis |\n\n
\n\n
\n\n## Tools\n\nArchitect Cert provides **18 MCP tools** that Claude uses to deliver the study experience:\n\n### Study Flow\n\n| Tool | Description |\n|------|-------------|\n| `start_assessment` | Begin with 15 diagnostic questions (lessons-first, one at a time) to determine your learning path |\n| `get_practice_question` | Get the next adaptive question with clickable A/B/C/D buttons (reviews > weak areas > new material) |\n| `submit_answer` | Grade your answer deterministically — presents interactive follow-up options |\n| `follow_up` | Handle post-answer actions: code examples, concept lessons, handouts, reference projects |\n| `get_section_details` | Deep dive into a specific task statement with full concept handout |\n\n### Progress & Planning\n\n| Tool | Description |\n|------|-------------|\n| `get_progress` | View overall study progress with mastery percentages per domain |\n| `get_curriculum` | Browse all 5 domains and 30 task statements with current mastery levels |\n| `get_weak_areas` | Identify topics that need the most work, ranked by weakness |\n| `get_study_plan` | Get personalized recommendations with multi-select domain focus |\n| `get_dashboard` | Open the visual progress dashboard in Claude Preview |\n\n### Practice Exams\n\n| Tool | Description |\n|------|-------------|\n| `start_practice_exam` | Take a full 60-question practice exam simulating the real certification |\n| `submit_exam_answer` | Submit and grade answers during a practice exam |\n| `get_exam_history` | View all past exam attempts with scores, trends, and per-domain comparison |\n\n### Capstone Build\n\n| Tool | Description |\n|------|-------------|\n| `start_capstone_build` | Start a guided capstone build — shape your project and validate criteria coverage |\n| `capstone_build_step` | Drive the capstone build: confirm, quiz, build, next, status, or abandon |\n| `capstone_build_status` | Check capstone build progress — current step, coverage, quiz performance |\n\n### Reference & Admin\n\n| Tool | Description |\n|------|-------------|\n| `scaffold_project` | Access reference projects for hands-on practice with real code |\n| `reset_progress` | Start over — requires explicit confirmation to prevent accidents |\n\nThe server also registers **8 interactive prompts** and **3 resource types** (concept handouts, reference projects, exam overview).\n\n
\n\n## Architecture\n\n```\n┌──────────────────────────────────────────────────────────────────────┐\n│ YOUR MACHINE │\n│ │\n│ ┌──────────────┐ ┌────────────────────────────────┐ │\n│ │ Claude Desktop│ │ Architect Cert MCP │ │\n│ │ Claude Code │◄─────►│ │ │\n│ │ Any MCP client│ stdio │ 18 tools │ │\n│ └──────┬───────┘ │ 8 prompts │ │\n│ │ │ 3 resource types │ │\n│ │ └──────────┬─────────────────────┘ │\n│ │ │ │\n│ │ ┌────────────────┼────────────────────┐ │\n│ │ │ │ │ │\n│ │ ~/.connectry- 390 questions 6 reference │\n│ │ architect/ 30 handouts projects │\n│ │ progress.db (bundled JSON/MD) (bundled TS) │\n│ │ │\n│ │ ┌──────────────────────────┐ │\n│ └──► Claude Preview (Dashboard) │ │\n│ │ localhost HTTP server │ │\n│ └──────────────────────────┘ │\n│ │\n└──────────────────────────────────────────────────────────────────────┘\n```\n\n### Core Components\n\n| Component | Technology | Purpose |\n|-----------|-----------|---------|\n| MCP Server | `@modelcontextprotocol/sdk` v1 | Registers tools, prompts, resources over stdio |\n| Grading Engine | Pure TypeScript functions | Deterministic answer verification |\n| Spaced Repetition | SM-2 algorithm | Optimal review scheduling |\n| Question Selector | Three-priority algorithm | Overdue reviews > weak areas > new material |\n| Follow-Up System | State-driven tool chain | Interactive post-answer detours |\n| Capstone Build Engine | 18-step interleaved builder | Guided learn-build-explain flow with LLM validation |\n| Dashboard Server | Node.js HTTP + HTML | Glassmorphism visual dashboard via Claude Preview |\n| Question Bank | 390 bundled JSON questions | Scenario-based, verified against docs |\n| Concept Handouts | 30 bundled markdown files | Structured study materials per task statement |\n| Reference Projects | 6 bundled TypeScript projects | Runnable code demonstrating each domain |\n| PDF Generator | Puppeteer + Marked | Branded handout PDFs for offline study |\n| Progress Store | `better-sqlite3` (WAL mode) | Persistent mastery, answers, schedules |\n\n### Interactive UI Architecture\n\nArchitect Cert doesn't build its own chat UI. Instead, it instructs Claude to use **built-in interactive tools**:\n\n- **AskUserQuestion** — Presents clickable buttons for A/B/C/D answers, follow-up actions, and domain selection. Supports single-select (radio buttons), multi-select (checkboxes), and free-text input via \"Other\".\n- **TodoWrite** — Creates visual progress checklists that update in real time during assessments, exams, and capstone builds.\n- **Claude Preview** — Renders the glassmorphism dashboard HTML on a local HTTP server.\n\nThis approach works across Claude Code and Claude Desktop without requiring any custom UI code on the client side.\n\n### Anti-Sycophancy Design\n\nThis server enforces honest grading at the protocol level — not just in prompts:\n\n1. **Deterministic grading** — `gradeAnswer()` is a pure function. No LLM is involved in judging correctness.\n2. **Tool-level enforcement** — The `submit_answer` tool description instructs Claude to relay results verbatim.\n3. **No partial credit** — Multiple choice, one correct answer. No \"you were on the right track.\"\n4. **Wrong answer explanations** — Every incorrect option has a specific `whyWrongMap` entry explaining the misconception.\n5. **System prompt rules** — Five anti-sycophancy directives prevent Claude from softening incorrect results.\n\n
\n\n## Question Bank Details\n\n| Metric | Value |\n|--------|-------|\n| Total questions | 390 |\n| Domains covered | 5 |\n| Task statements covered | 30 |\n| Questions per task statement | 13 |\n| Difficulty distribution | ~4 easy, 5 medium, ~4 hard per task |\n| Answer key balance | Distributed across A/B/C/D |\n| Question format | Scenario-based multiple choice |\n| Each question includes | Scenario, question, 4 options, explanation, why-wrong-map, references |\n| Source material | Anthropic official documentation |\n\n
\n\n## Data Storage\n\n- Progress is stored locally at `~/.connectry-architect/progress.db` (SQLite, WAL mode)\n- Your user config lives at `~/.connectry-architect/config.json` (auto-created on first run)\n- No cloud, no accounts, no telemetry — everything stays on your machine\n\n
\n\n## Contributing\n\nWe welcome contributions! Here's how to get started:\n\n```bash\n# Clone the repo\ngit clone https://github.com/Connectry-io/connectrylab-architect-cert-mcp.git\ncd connectrylab-architect-cert-mcp\n\n# Install dependencies\nnpm install\n\n# Build\nnpm run build\n\n# Run tests\nnpm test\n\n# Generate PDF handouts\nnpm run generate:pdfs\n\n# Run locally\nnode dist/index.js\n```\n\n### Project Structure\n\n```\nsrc/\n├── index.ts # MCP server entry point\n├── config.ts # User config management\n├── types.ts # All TypeScript interfaces\n├── data/\n│ ├── loader.ts # Lazy-cached data loading\n│ ├── curriculum.json # 30 task statements\n│ ├── questions/ # 390 questions (5 domain files)\n│ ├── handouts/ # 30 concept handouts (markdown)\n│ ├── criteria.ts # 30 task statement criteria for capstone validation\n│ ├── build-steps.ts # 18 capstone build step definitions\n│ └── system-prompt.ts # Anti-sycophancy rules\n├── db/\n│ ├── schema.ts # SQLite schema (9 tables)\n│ ├── store.ts # Database initialization\n│ ├── mastery.ts # Mastery level calculations\n│ ├── answers.ts # Answer recording\n│ ├── review-schedule.ts # SM-2 review scheduling\n│ ├── capstone.ts # Capstone build CRUD operations\n│ ├── users.ts # User management\n│ └── exam-attempts.ts # Practice exam tracking\n├── engine/\n│ ├── grading.ts # Deterministic grading\n│ ├── spaced-repetition.ts # SM-2 algorithm\n│ ├── question-selector.ts # Priority-based selection\n│ ├── exam-builder.ts # Practice exam generation\n│ └── adaptive-path.ts # Learning path recommendations\n├── tools/ # 18 MCP tool handlers\n│ ├── index.ts # Tool registration\n│ ├── elicit.ts # MCP elicitation helper (graceful fallback)\n│ ├── start-assessment.ts\n│ ├── submit-answer.ts\n│ ├── get-practice-question.ts\n│ ├── follow-up.ts\n│ ├── get-progress.ts\n│ ├── get-curriculum.ts\n│ ├── get-section-details.ts\n│ ├── get-weak-areas.ts\n│ ├── get-study-plan.ts\n│ ├── start-practice-exam.ts\n│ ├── submit-exam-answer.ts\n│ ├── get-exam-history.ts\n│ ├── scaffold-project.ts\n│ ├── start-capstone-build.ts\n│ ├── capstone-build-step.ts\n│ ├── capstone-build-status.ts\n│ ├── reset-progress.ts\n│ └── dashboard.ts\n├── ui/\n│ ├── server.ts # Dashboard HTTP server\n│ ├── dashboard.html # Glassmorphism dashboard UI\n│ ├── meta.ts # Quiz widget metadata\n│ └── loader.ts # HTML asset loader\n├── prompts/ # 8 MCP prompt definitions\n└── resources/ # 3 MCP resource types\n\nprojects/\n├── capstone/ # All 5 domains — multi-agent support system\n├── d1-agentic/ # Domain 1 — agentic loop research coordinator\n├── d2-tools/ # Domain 2 — MCP server with tool patterns\n├── d3-config/ # Domain 3 — Claude Code configuration layout\n├── d4-prompts/ # Domain 4 — extraction & prompt engineering\n└── d5-context/ # Domain 5 — context management & reliability\n\nscripts/\n└── generate-pdfs.ts # PDF handout generator\n```\n\n
\n\n## License\n\nMIT © [Connectry Labs](https://connectry.io/labs)\n\n
\n\n## Credits\n\n- [Anthropic](https://anthropic.com) — Claude & the Claude Certified Architect certification program\n- [Model Context Protocol](https://modelcontextprotocol.io/) — The protocol that makes this possible\n- [better-sqlite3](https://github.com/WiseLibs/better-sqlite3) — Fast, synchronous SQLite for Node.js\n\n
\n\n---\n\n

\n Built with care by Connectry Labs\n

\n\n

\n GitHub •\n npm •\n Architect Cert •\n Connectry Labs\n

\n", "bytes": 36205, "sha": "7cc4d79bbdefeb08f1abce39e6451e35963f53afaf9c7fa87548299a5e948a06", "repo_slug": "connectry-io/connectrylab-architect-cert-mcp", "fonte": "repo", "truncated": false, "api": "https://agentalog.com/api/listings/mcp_io_github_weareconnectry_connectry_archi_831ca4a8/readme" }