Templates
50 built-in templates covering common consulting slide types. Use with POST /v1/renderfor instant, deterministic rendering (<1s, $0.03-$0.05).
Template Categories
structure (16)
Data Table, Org Chart, Risk Matrix
process (8)
Timeline, Chevrons, Roadmap, Funnel
metrics (6)
KPI Dashboard, TAM/SAM/SOM
narrative (6)
Executive Summary, Bullet List
comparison (5)
SWOT, Feature Comparison, Pricing
chart (4)
Waterfall, Bar, Stacked Bar
closing (3)
Key Takeaway, Next Steps
title (2)
Title Slide, Section Divider
GET /v1/templates
List all templates. Filter by category, audience, or style.
# List all templates curl https://api.slideforge.dev/v1/templates \ -H "Authorization: Bearer sf_live_YOUR_KEY" # Filter by category curl "https://api.slideforge.dev/v1/templates?category=metrics" \ -H "Authorization: Bearer sf_live_YOUR_KEY"
POST /v1/templates/suggest
Find the best template for a natural language brief. Returns ranked suggestions with has_good_match indicating if a template fits well enough for rendering (vs creative generation).
# Find best template for a brief
curl -X POST https://api.slideforge.dev/v1/templates/suggest \
-H "Authorization: Bearer sf_live_YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{"brief": "4 KPI metrics: Revenue, Clients, Margin, NPS"}'{
"suggestions": [
{
"template_id": "cbddc7b0-...",
"name": "KPI Dashboard",
"description": "Grid of 2-6 KPI metrics...",
"category": "metrics",
"score": 0.081,
"recommended": true,
"schema": { ... }
}
],
"has_good_match": true
}If has_good_match: true, use template_id with POST /v1/render. Otherwise, use POST /v1/generate.
POST /v1/templates/search
Semantic search across the template library. Uses hybrid vector + keyword + semantic reranking. Returns ranked results with descriptions and param schemas.
# Semantic search across templates
curl -X POST https://api.slideforge.dev/v1/templates/search \
-H "Authorization: Bearer sf_live_YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{"query": "compare multiple options side by side with ratings"}'Tip: auto strategy
When using POST /v1/generate, the default strategy="auto"automatically checks templates first. If a good match exists, it renders instantly instead of running AI generation. You don't need to call suggest manually in most cases.