43-spark ? EVA Spark Springboard

EVA Ecosystem Integration

Tool	Purpose	How to Use
37-data-model	Single source of truth for all project entities	GET http://localhost:8010/model/projects/43-spark
29-foundry	Agentic capabilities (search, RAG, eval, observability)	C:\eva-foundry\eva-foundation\29-foundry
48-eva-veritas	Trust score and coverage audit	MCP tool: audit_repo / get_trust_score
07-foundation-layer	Copilot instructions primer + governance templates	MCP tool: apply_primer / audit_project

Agent rule: Query the data model API before reading source files.

Invoke-RestMethod "http://localhost:8010/model/agent-guide"   # complete protocol
Invoke-RestMethod "http://localhost:8010/model/agent-summary" # all layer counts

---

Purpose: Reusable launchpad for building and iterating on EVA UI/UX frontends using GitHub Spark.
Current target: EVA-JP v1.2 Frontend Rebuild (see 44-eva-jp-spark/).
Pattern: Draft prompt here ? refine ? paste into Spark ? iterate in Codespace with agent mode.

---

What Is This Folder?

43-spark is a shared design system and prompt engineering workspace for all EVA UI/UX projects that use GitHub Spark as the scaffolding vehicle. It contains:

Path	Purpose
README.md	This file ? orientation and quick-start
PLAN.md	Phased delivery plan and Spark iteration strategy
STATUS.md	Live tracking board ? what is done, in progress, blocked
ACCEPTANCE.md	Master acceptance criteria used across all EVA Spark apps
.github/copilot-instructions.md	Repository-level Copilot instructions (agent mode context)
.copilot/skills/eva-ui.md	Skill: @eva/ui + @eva/gc-design-system component patterns and GC token rules
.copilot/skills/a11y-wcag.md	Skill: WCAG 2.1 AA rules enforced in every EVA app
.copilot/skills/i18n-enfr.md	Skill: EN/FR internationalisation with i18next
.copilot/skills/spark-workflow.md	Skill: How to structure Spark prompts and iterate
workflows/spark-prompt-workflow.md	Step-by-step workflow from requirement ? final Spark prompt
prompts/FINAL-SPARK-PROMPT.md	The consolidated, ready-to-paste Spark prompt for EVA-JP

---

Quick Start

1 ? Paste the initial Spark prompt

Copy everything inside the first fenced code block in prompts/FINAL-SPARK-PROMPT.md and paste it into github.com/spark.

Spark will generate a React + TypeScript app with a live preview. This consumes 4 premium Copilot requests.

2 ? Review the preview

Check the live preview for:

• Header renders with logo, nav, language toggle
• Chat screen is the default route with a streaming input area
• No visible v8 Fluent component artefacts

3 ? Open in Codespace

Click "Open in Codespace" in Spark. Once the Codespace starts:

npm run dev

The app will be served at localhost:5173.

4 ? Attach context and iterate

In VS Code inside the Codespace, open Copilot Chat in agent mode and attach:

• eva-jp-rebuild-spec.json (from 44-eva-jp-spark/)
• PRD-REBUILD-FLUENT-V9.md (from 44-eva-jp-spark/)

Then paste each ITERATION PROMPT block from prompts/FINAL-SPARK-PROMPT.md in sequence.

5 ? Validate

Run the acceptance checks in ACCEPTANCE.md. All AC-01 through AC-15 must pass before merging.

---

How Spark Works (Key Facts for Prompting)

Fact	Impact on prompting
Stack is React + TypeScript (opinionated, cannot be changed)	Aligns perfectly with EVA; no adaptation needed
UI layer must use @eva/ui + @eva/gc-design-system	Import EvaXxx wrappers first; raw @fluentui/react-components only for un-wrapped primitives
External libraries are allowed but not guaranteed compatible	List exact versions; test after each library add
Each natural-language prompt costs 4 premium Copilot requests	Use iteration prompts sparingly; batch related work
Organisation sharing available	Publish spark to your GitHub org for team review
Two-way sync with a GitHub repo	Create a repo immediately after first generation; keep Codespace changes in sync
Agent mode in Codespace ? Spark prompts	Agent mode is for surgical edits; Spark prompts scaffold structure
Spark does not know your internal APIs	All API paths must be explicit in the prompt (e.g. /api/userGroupInfo)

---

Folder Conventions

43-spark/
??? .github/
?   ??? copilot-instructions.md   ? Agent mode context for this repo
??? .copilot/
?   ??? skills/
?       ??? fluent-v9.md
?       ??? a11y-wcag.md
?       ??? i18n-enfr.md
?       ??? spark-workflow.md
??? workflows/
?   ??? spark-prompt-workflow.md  ? Human process guide
??? prompts/
?   ??? FINAL-SPARK-PROMPT.md     ? Ready-to-paste prompt (single source of truth)
??? README.md
??? PLAN.md
??? STATUS.md
??? ACCEPTANCE.md

---

Related Projects

Folder	Description
44-eva-jp-spark/	EVA-JP v1.2 Spark output + spec files ? github.com/eva-foundry/44-eva-jp-spark
31-eva-faces/	Design specs, ADO work items, PRD documents
40-eva-devbench/	Local dev server for EVA frontend testing

---

? What Is Next (February 2026)

Phase 0 is complete. The standalone repo https://github.com/eva-foundry/eva-jp-spark has been created.

Immediate next steps:

1. cd 44-eva-jp-spark && npm install ? install all runtime + dev dependencies
2. Open github.com/spark, paste ITERATION PROMPT 1 from prompts/FINAL-SPARK-PROMPT.md
3. Review the live preview (Layout shell, language toggle, skip link)
4. Open Codespace ? attach eva-jp-rebuild-spec.json ? paste Iteration 2, 3, 4, 5 in sequence
5. After each iteration, run the AC checks in ACCEPTANCE.md
6. Add .github/workflows/ci.yml to the eva-foundry/44-eva-jp-spark repo for build + typecheck gate

---

Key References

• GitHub Spark docs
• @eva/ui ? EVA shared component library
• @eva/gc-design-system ? GC theme + tokens
• Fluent UI React v9 (underlying primitive library; use @eva wrappers first)
• PubSec-Info-Assistant frontend ? API contract source
• WCAG 2.1 AA ? accessibility standard
• GC Standard on Web Accessibility