RAG Frameworks · archie0732

healthy-diet-ai-agent

Healthy Diet AI Agent is a TypeScript backend service for nutrition guidance, food image analysis, and document-based knowledge retrieval. It runs standalone with SQLite or integrates with Supabase, supporting both HTTP API and CLI deployment modes.

Talk to DEV.co Discuss Implementation

Source: GitHub — github.com/archie0732/healthy-diet-ai-agent

618

GitHub stars

Forks

TypeScript

Primary language

MIT

License (OSI-approved)

Key facts

Objective fields from the source. Values we can't verify are shown as “Unknown” rather than guessed.

Field	Value
Repository	archie0732/healthy-diet-ai-agent
Owner	archie0732
Primary language	TypeScript
License	MIT — OSI-approved
Stars	618
Forks	15
Open issues	5
Latest release	Unknown
Last updated	2026-07-05
Source	https://github.com/archie0732/healthy-diet-ai-agent

What healthy-diet-ai-agent is

Node.js/Bun runtime with Express, LangChain/LangGraph agentic framework, OpenAI-compatible LLM routing, RAG via document embedding, SQLite or Supabase storage, and Docker Compose orchestration. Supports multimodal input (text, images) and knowledge graph extraction.

Quickstart

Get the healthy-diet-ai-agent source

Clone the repository and explore it locally.

terminalbash

1git clone https://github.com/archie0732/healthy-diet-ai-agent.git2cd healthy-diet-ai-agent3# follow the project's README for install & configuration

Need it deployed, integrated, or customized instead? DEV.co ships production installs.

Best use cases

Self-hosted nutrition chatbot

Deploy standalone with SQLite for a privacy-preserving dietary advice service without external database dependencies. Suitable for clinics, wellness apps, or internal health platforms.

Document-grounded health knowledge system

Ingest dietary guidelines, medical documents, or organizational nutrition data via RAG to ground agent responses. Minimizes hallucination on nutrition-critical advice.

Meal analysis workflow integration

Leverage image analysis capability within existing health ecosystems (via Supabase mode) to interpret food photos and correlate with user profiles and historical records.

Implementation considerations

Requires OpenAI-compatible LLM endpoint (or Google Gemini). API key and base URL must be provided; no bundled model or offline fallback.
SQLite mode uses file-based storage (./data/healthy-diet-agent.db); plan for backup, replication, and data durability in production.
RAG pipeline expects documents ingested as markdown in knowledge_base/ingested_markdown/. Manual preprocessing may be needed for PDFs or structured data.
Agent behavior is configurable via agent_config.json; no UI for model tuning, so changes require code deployment.
Docker Compose available but requires explicit Bun runtime and manual schema bootstrap for Supabase mode; verify schema compatibility before scaling.

When to avoid it — and what to weigh

High-volume production with strict SLA — No release versioning, no performance benchmarks, and no documented uptime guarantees. Active development (last push July 2026) may introduce breaking changes.
Regulated medical/clinical deployment without audit trail — SQLite default lacks enterprise audit logging, encryption-at-rest, or compliance features required for HIPAA, GDPR, or similar. Would require custom hardening.
Requires guaranteed LLM model stability — Agent behavior depends on upstream OpenAI or Google Gemini models and API compatibility. No fallback strategy or model versioning documented; breaking API changes may require code updates.
Multi-tenant SaaS with isolated data — Supabase integration exists, but no multi-tenant row-level security (RLS) or isolation patterns are documented. Custom implementation required.

License & commercial use

MIT License. Permissive OSI license allowing commercial use, modification, and distribution with attribution. No patent clauses or restrictions on use case.

MIT permits commercial use without additional licensing fees or permissions. However, upstream LLM APIs (OpenAI, Google Gemini) and Supabase require separate commercial agreements. Ensure your AI provider's terms permit production use for health/nutrition advice.

DEV.co evaluation signals

Editorial assessment — not user reviews. Directional, with an explicit confidence level.

Signal	Assessment
Maintenance	Active
Documentation	Adequate
License clarity	Clear
Deployment complexity	Moderate
DEV.co fit	Good
Assessment confidence	High

Security considerations

Standalone SQLite stores all data unencrypted on disk; no encryption-at-rest or secrets management integrated. API keys (AI_API_URL, API keys) passed via .env with no rotation mechanism. Supabase mode delegates auth/encryption to Supabase. No documented input validation or injection protection for RAG queries or image uploads. No rate limiting or DDoS mitigation. Recommend adding secrets manager integration, request validation middleware, and API key rotation before production health data exposure.

Alternatives to consider

LangChain-only (no pre-built agent)

Lower-level control, but requires building agent orchestration, storage adapters, and HTTP server from scratch. Steeper implementation cost.

Existing health API platforms (Nutritionix, USDA FoodData Central)

Pre-built nutrition databases and APIs, but no conversational agent or RAG. Requires integration glue; better for data lookup than advice generation.

Claude AI API with custom RAG (e.g., via Langsmith)

Anthropic models often score higher on long-context reasoning; Langsmith provides observability. Requires separate architecture; less opinionated than this project.

Software development agency

Build on healthy-diet-ai-agent with DEV.co software developers

Evaluate Healthy Diet AI Agent for your next health or wellness project. Start with SQLite standalone mode, integrate your own documents, and scale to Supabase as needed.

Talk to DEV.co

Related open-source tools

Surfaced by semantic similarity across the DEV.co open-source index.

agentset OmAgent RagaAI-Catalyst

Related on DEV.co

Explore the category and the services that help you build with it.

RAG Frameworks AI Application Development API Development Cloud Deployment Talk to DEV.co

healthy-diet-ai-agent FAQ

Can I use this in production without Supabase?

Yes. SQLite standalone mode supports production, but you must handle backups, replication, and security hardening yourself. Suitable for single-instance deployments; not for distributed systems.

What LLM models are supported?

OpenAI-compatible APIs (GPT-4, GPT-3.5) and Google Gemini. Must provide AI_API_URL and API key via .env. No bundled model; requires external LLM service.

How do I add my own nutrition guidelines?

Place documents in knowledge_base/ingested_markdown/ and use the /api/documents endpoints to index them. NUTRITION_RULES.md serves as the ground-truth reference; edit it and re-ingest to update agent behavior.

Is this HIPAA-compliant?

Not out-of-the-box. SQLite mode has no audit logging or encryption-at-rest. Supabase mode depends on Supabase's compliance features. Custom hardening required; not recommended for PHI without professional security review.

Software developers & web developers for hire

From first prototype to production, DEV.co delivers software development services around tools like healthy-diet-ai-agent. Our software development agency staffs experienced software developers and web developers for custom software development, web development, integrations, and ongoing support across rag frameworks and beyond.

Ready to build a nutrition AI service?

Evaluate Healthy Diet AI Agent for your next health or wellness project. Start with SQLite standalone mode, integrate your own documents, and scale to Supabase as needed.

Talk to DEV.co Schedule a Consultation