Skip to content

Phase 2: Templates + Agent Catalog

Goal

Expand from 3 to 15+ agents. Build out all template types. Enable community contributions.

New templates

template-agent-translator

  • NLLB-200 (200 languages)
  • Language detection
  • Side-by-side input/output
  • Clipboard integration

template-agent-summarizer

  • BART-small or T5-small
  • URL fetcher (via CORS proxy) or text paste
  • Adjustable length
  • Key points extraction

template-agent-code

  • WebContainers for tooling (ESLint, tsc, Prettier)
  • Optional Ollama for LLM review
  • Syntax highlighting (CodeMirror)
  • Multi-file support

template-agent-chat

  • WebLLM (Llama 3.2 1B or Phi-3 mini)
  • Conversation UI
  • System prompt customization
  • Optional Ollama backend

template-agent-ocr

  • Florence-2 or Tesseract.js
  • Camera capture + image upload
  • Structured output (JSON, text, table)
  • PDF support via PDF.js

template-agent-diffusion

  • Stable Diffusion (WebGPU)
  • Text prompt input
  • Generation progress
  • Gallery of results
  • Download/share

template-agent-embeddings

  • all-MiniLM-L6 (23MB)
  • Upload documents → build local search index
  • Semantic search across documents
  • IndexedDB persistence

template-agent-data

  • CSV/JSON parser
  • Chart.js for visualization
  • Small LLM for natural language queries
  • Export results

New agents (built from templates)

Agent Template Unique features
Translator agent-translator Saved translations history
Article Summarizer agent-summarizer URL input, RSS feed support
Code Linter agent-code ESLint + Biome profiles
Local ChatBot agent-chat System prompt library
Receipt Scanner agent-ocr Expense categorization
Image Creator agent-diffusion Prompt suggestions
Document Search agent-embeddings Multi-document support
CSV Analyzer agent-data Auto-chart suggestions
Sentiment Tracker Custom (DistilBERT) Social media monitoring
Flashcard Maker Custom (T5) Spaced repetition
Font Previewer Custom (no model) Google Fonts integration
Color Palette Custom (CLIP) Image-to-palette

Community contribution flow

  1. Developer runs fags init my-agent --template agent-whisper
  2. Builds their agent locally
  3. Runs fags check (compliance)
  4. Runs fags publish (creates repo, provisions hosting)
  5. Agent appears on store after admin approval

Shared model cache (Phase 2 feature)

Service Worker at the store level that caches models across agents:

User opens Transcriber → downloads Whisper (240MB)
User opens Meeting Notes → Whisper already cached!
User opens Podcast Summarizer → Whisper already cached!

Reduces bandwidth and improves UX across the agent catalog.

Definition of done (Phase 2)

  • [ ] 8 templates available
  • [ ] 15+ agents on store
  • [ ] Shared model cache working
  • [ ] Community publish flow documented
  • [ ] VibeCode agent can build new agents from description
  • [ ] Categories and search on store site