Agentic AI Atlasby a5c.ai
OverviewWikiGraphFor AgentsEdgesSearchWorkspace
/
GitHubDocsDiscord
iiRecord
Agentic AI Atlas · yizhiyanhua-ai/fireworks-tech-graph
page:docs-reference-repos-yizhiyanhua-ai-fireworks-tech-graph-researcha5c.ai
Search record views/
Record · tabs

Available views

II.Record viewspp. 1 - 1
overviewarticlejsongraph
II.
Page JSON

page:docs-reference-repos-yizhiyanhua-ai-fireworks-tech-graph-research

Structured · live

yizhiyanhua-ai/fireworks-tech-graph json

Inspect the normalized record payload exactly as the atlas UI reads it.

File · wiki/docs/reference-repos/yizhiyanhua-ai/fireworks-tech-graph/research.mdCluster · wiki
Record JSON
{
  "id": "page:docs-reference-repos-yizhiyanhua-ai-fireworks-tech-graph-research",
  "_kind": "Page",
  "_file": "wiki/docs/reference-repos/yizhiyanhua-ai/fireworks-tech-graph/research.md",
  "_cluster": "wiki",
  "attributes": {
    "nodeKind": "Page",
    "sourcePath": "docs/reference-repos/yizhiyanhua-ai/fireworks-tech-graph/research.md",
    "sourceKind": "repo-docs",
    "title": "yizhiyanhua-ai/fireworks-tech-graph",
    "displayName": "yizhiyanhua-ai/fireworks-tech-graph",
    "slug": "docs/reference-repos/yizhiyanhua-ai/fireworks-tech-graph/research",
    "articlePath": "wiki/docs/reference-repos/yizhiyanhua-ai/fireworks-tech-graph/research.md",
    "article": "\n# yizhiyanhua-ai/fireworks-tech-graph\n\n- **Archetype**: utility-with-skill\n- **Stars**: 1,512\n- **Last pushed**: 2026-04-12\n- **License**: MIT\n- **Discovered**: 2026-04-12\n- **Skills found**: 1\n\n## Summary\nClaude Code skill for generating production-quality SVG+PNG technical diagrams. Supports 8 diagram types and 5 visual styles with deep AI/Agent domain knowledge. Uses rsvg-convert for SVG-to-PNG export. Includes 4 helper scripts: generate-diagram.sh (validate+export), generate-from-template.py (template-based SVG creation), validate-svg.sh (syntax validation), and test-all-styles.sh (batch testing). Bilingual triggers (Chinese + English).\n\n## Assessment\nMEDIUM VALUE. The helper script architecture is well-designed: a Python template renderer produces SVGs from structured JSON input (title, nodes, arrows, legend), a bash validator checks XML syntax/tag balance/marker references/path data, and a test script batch-validates all styles. This validation-centric approach to SVG generation (generate -> validate -> export) is a reusable pattern for any code-generation skill. The template-based generation avoids the fragility of raw SVG string construction. The bilingual trigger pattern shows how to make skills accessible across language communities.\n\n## Extraction Priority\n- Low-Medium\n- Rationale: The diagram generation procedure and validation pipeline are useful patterns, but the skill is primarily a rendering tool. The template-based generation + validation loop is the most transferable element.\n\n## Processes\n- **Validated Diagram Generation Process**: Select diagram type + style -> generate from template with structured JSON -> validate SVG (XML syntax, tag balance, markers, paths, attributes) -> export PNG. The validation step prevents malformed output.\n- **Batch Style Testing**: Generate diagrams across multiple sizes and styles -> validate all -> produce test report. A CI-like quality gate for visual output.\n\n## Plugin Ideas\n- **Tech Diagram Generator plugin**: Install.md-driven plugin that adds diagram generation skills with template-based SVG creation and validation. Configurable diagram types and visual styles.\n\n## Patterns\n- **Template-based generation with validation**: Generate from structured input (JSON) via templates rather than raw string construction. Validate output before delivery. Prevents fragile code generation.\n- **Helper script decomposition**: Four focused scripts (generate, template-render, validate, batch-test) each handling one concern. The skill orchestrates them as a pipeline.\n- **Bilingual triggers**: Both Chinese and English trigger phrases in the skill description. Simple internationalization for skill discovery.\n- **SVG validation checklist**: XML syntax, tag balance, marker references, attribute completeness, path data integrity. A reusable validation pattern for any structured output format.\n\n## Library Mapping\n\n| Extractable Process | Library Status | Action | Existing Path | Target Placement |\n|-------------------|----------------|--------|---------------|------------------|\n| Validated Diagram Generation Process | NEW | Template-based SVG creation with comprehensive validation and PNG export | - | specializations/shared/validated-diagram-generation.js |\n| Batch Style Testing | NEW | CI-like quality gate for visual output across multiple styles and sizes | - | specializations/shared/batch-style-testing.js |\n| Template-Based Generation with Validation | NEW | Structured input generation via templates with output validation to prevent fragility | - | specializations/shared/template-based-generation-validation.js |\n| Bilingual Triggers | NEW | Internationalization pattern with both Chinese and English trigger phrases | - | specializations/shared/bilingual-triggers.js |\n| SVG Validation Checklist | NEW | Comprehensive validation pattern for structured output formats | - | specializations/shared/svg-validation-checklist.js |\n\n## Plugin Marketplace Mapping\n\n| Plugin Idea | Marketplace Status | Action | Existing Plugin | Target Placement |\n|-------------|-------------------|--------|-----------------|------------------|\n| Tech Diagram Generator Plugin | NEW | Template-based diagram generation with validation and configurable styles | - | plugins/a5c/marketplace/blueprints/tech-diagram-generator/ |\n",
    "documents": []
  },
  "outgoingEdges": [],
  "incomingEdges": [
    {
      "from": "page:docs-reference-repos",
      "to": "page:docs-reference-repos-yizhiyanhua-ai-fireworks-tech-graph-research",
      "kind": "contains_page"
    }
  ]
}

Shortcuts

Back to overview
Open graph tab