Stitch Build Loop

You are an autonomous frontend builder participating in an iterative site-building loop. Your goal is to generate a page using Stitch, integrate it into the site, and prepare instructions for the next iteration.

Overview

The Build Loop pattern enables continuous, autonomous website development through a "baton" system. Each iteration:

1. Reads the current task from a baton file (next-prompt.md)
2. Generates a page using Stitch MCP tools
3. Integrates the page into the site structure
4. Writes the next task to the baton file for the next iteration

Prerequisites

Required:

• Access to the Stitch MCP Server
• A Stitch project (existing or will be created)
• A DESIGN.md file (generate one using the design-md skill if needed)
• A SITE.md file documenting the site vision and roadmap

Optional:

• Chrome DevTools MCP Server — enables visual verification of generated pages
• Qdrant Memory Skill — enables semantic context retrieval and long-term memory

The Baton System

The next-prompt.md file acts as a relay baton between iterations:

---
page: about
---

A page describing how jules.top tracking works.

**DESIGN SYSTEM (REQUIRED):**
[Copy from DESIGN.md Section 6]

**Page Structure:**

1. Header with navigation
2. Explanation of tracking methodology
3. Footer with links

Critical rules:

• The page field in YAML frontmatter determines the output filename
• The prompt content must include the design system block from DESIGN.md
• You MUST update this file before completing your work to continue the loop

Execution Protocol

Step 1: Read the Baton

Parse next-prompt.md to extract:

• Page name from the page frontmatter field
• Prompt content from the markdown body

Step 2: Check Context Files

Before generating, read these files:

Step 2.5: Check Memory (Optional)

If the Qdrant Memory Skill is available, retrieve relevant context to ensure consistency:

1. Retrieve Context: Call qdrant_search_memory (or retrieve_context script) with technical queries related to the current task.
   • Example query: "navigation structure", "footer design", "auth flow pattern"
   • Filter by project if applicable.
2. Check Decisions: Look for past architectural decisions (type: "decision") that might impact this page.

Important checks:

• Section 4 (Sitemap) — Do NOT recreate pages that already exist
• Section 5 (Roadmap) — Pick tasks from here if backlog exists
• Section 6 (Creative Freedom) — Ideas for new pages if roadmap is empty

Step 3: Generate with Stitch

Use the Stitch MCP tools to generate the page:

1. Discover namespace: Run list_tools to find the Stitch MCP prefix
2. Get or create project:
   • If stitch.json exists, use the projectId from it
   • Otherwise, call [prefix]:create_project and save the ID to stitch.json
3. Generate screen: Call [prefix]:generate_screen_from_text with:
   • projectId: The project ID
   • prompt: The full prompt from the baton (including design system block)
   • deviceType: DESKTOP (or as specified)
4. Retrieve assets: Call [prefix]:get_screen to get:
   • htmlCode.downloadUrl — Download and save as queue/{page}.html
   • screenshot.downloadUrl — Download and save as queue/{page}.png

Step 4: Integrate into Site

1. Move generated HTML from queue/{page}.html to site/public/{page}.html
2. Fix any asset paths to be relative to the public folder
3. Update navigation:
   • Find existing placeholder links (e.g., href="#") and wire them to the new page
   • Add the new page to the global navigation if appropriate
4. Ensure consistent headers/footers across all pages

Step 4.5: Visual Verification (Optional)

If the Chrome DevTools MCP Server is available, verify the generated page:

1. Check availability: Run list_tools to see if chrome* tools are present
2. Start dev server: Use Bash to start a local server (e.g., npx serve site/public)
3. Navigate to page: Call [chrome_prefix]:navigate to open http://localhost:3000/{page}.html
4. Capture screenshot: Call [chrome_prefix]:screenshot to capture the rendered page
5. Visual comparison: Compare against the Stitch screenshot (queue/{page}.png) for fidelity
6. Stop server: Terminate the dev server process

Note: This step is optional. If Chrome DevTools MCP is not installed, skip to Step 5.

Step 5: Update Site Documentation

Modify SITE.md:

• Add the new page to Section 4 (Sitemap) with [x]
• Remove any idea you consumed from Section 6 (Creative Freedom)
• Update Section 5 (Roadmap) if you completed a backlog item

Step 5.5: Store Memory (Optional)

If Qdrant Memory Skill is available:

1. Store Completion: Use qdrant_store_memory to record the completion of the page.
   • Content: "Completed page: {page}. Implemented features: [list features]."
   • Metadata: type: "milestone", page: "{page}"
2. Store Decisions: If you made any structural or design decisions (e.g., "Standardized footer links"), store them:
   • Metadata: type: "decision", topic: "navigation"

Step 6: Prepare the Next Baton (Critical)

You MUST update next-prompt.md before completing. This keeps the loop alive.

1. Decide the next page:
   • Check SITE.md Section 5 (Roadmap) for pending items
   • If empty, pick from Section 6 (Creative Freedom)
   • Or invent something new that fits the site vision
2. Write the baton with proper YAML frontmatter:

---
page: achievements
---

A competitive achievements page showing developer badges and milestones.

**DESIGN SYSTEM (REQUIRED):**
[Copy the entire design system block from DESIGN.md]

**Page Structure:**

1. Header with title and navigation
2. Badge grid showing unlocked/locked states
3. Progress bars for milestone tracking

File Structure Reference

project/
├── next-prompt.md      # The baton — current task
├── stitch.json         # Stitch project ID (persist this!)
├── DESIGN.md           # Visual design system (from design-md skill)
├── SITE.md             # Site vision, sitemap, roadmap
├── queue/              # Staging area for Stitch output
│   ├── {page}.html
│   └── {page}.png
└── site/public/        # Production pages
    ├── index.html
    └── {page}.html

Orchestration Options

The loop can be driven by different orchestration layers:

The skill is orchestration-agnostic — focus on the pattern, not the trigger mechanism.

Design System Integration

This skill works best with the design-md skill:

1. First time setup: Generate DESIGN.md using the design-md skill from an existing Stitch screen
2. Every iteration: Copy Section 6 ("Design System Notes for Stitch Generation") into your baton prompt
3. Consistency: All generated pages will share the same visual language

Common Pitfalls

• ❌ Forgetting to update next-prompt.md (breaks the loop)
• ❌ Recreating a page that already exists in the sitemap
• ❌ Not including the design system block in the prompt
• ❌ Leaving placeholder links (href="#") instead of wiring real navigation
• ❌ Forgetting to persist stitch.json after creating a new project

Troubleshooting

AGI Framework Integration

Qdrant Memory Integration

Before executing complex tasks with this skill:

python3 execution/memory_manager.py auto --query "<task summary>"

Decision Tree:

• Cache hit? Use cached response directly — no need to re-process.
• Memory match? Inject context_chunks into your reasoning.
• No match? Proceed normally, then store results:

python3 execution/memory_manager.py store \
  --content "Description of what was decided/solved" \
  --type decision \
  --tags stitch-loop <relevant-tags>

Note: Storing automatically updates both Vector (Qdrant) and Keyword (BM25) indices.

Agent Team Collaboration

• Strategy: This skill communicates via the shared memory system.
• Orchestration: Invoked by orchestrator via intelligent routing.
• Context Sharing: Always read previous agent outputs from memory before starting.

Local LLM Support

When available, use local Ollama models for embedding and lightweight inference:

• Embeddings: nomic-embed-text via Qdrant memory system
• Lightweight analysis: Local models reduce API costs for repetitive patterns