enki-runtime 0.1.4

# RAG Agent Mesh Configuration for Enki Runtime

#

# This file demonstrates how to configure RAG agents with memory backends using TOML.

# Load with: MeshConfig::from_file("rag_agents.toml")



name = "rag_mesh"



# Shared memory configuration for all agents (default fallback)

# Backend options: "inmemory" (default), "sqlite", "redis"

[memory]

backend = "inmemory"

max_entries = 1000

ttl_seconds = 3600



# RAG Agent - retrieves context and generates grounded responses

# Uses the shared mesh [memory] configuration above (no [agents.memory] specified)

[[agents]]

name = "rag_agent"

model = "ollama::gemma3:latest"

system_prompt = """
You are a helpful assistant that answers questions based on the provided context. 
Always ground your answers in the given context. If the context doesn't contain 
enough information, say so. Be concise and informative.
"""

temperature = 0.3

max_tokens = 512



# Knowledge Curator - has its OWN separate memory configuration

# This demonstrates per-agent memory isolation

[[agents]]

name = "curator"

model = "ollama::gemma3:latest"

system_prompt = """
You are a knowledge curator. Your job is to organize and summarize information 
into well-structured knowledge entries. Extract key facts and organize them 
clearly and concisely.
"""

temperature = 0.2

max_tokens = 1024



# Per-agent memory configuration (overrides mesh default)

[agents.memory]

backend = "inmemory"

max_entries = 500    # Smaller limit for curator's focused storage

ttl_seconds = 7200   # Longer TTL for curated knowledge



# Persistent Agent Example - uses SQLite for persistent storage

# Uncomment to use (requires 'sqlite' feature)

# [[agents]]

# name = "persistent_agent"

# model = "ollama::gemma3:latest"

# system_prompt = "You are a persistent agent with SQLite storage."

# 

# [agents.memory]

# backend = "sqlite"

# path = "./persistent_agent_memory.db"



# Redis Agent Example - uses Redis for distributed storage

# Uncomment to use (requires 'redis' feature)

# [[agents]]

# name = "distributed_agent"

# model = "ollama::gemma3:latest"

# system_prompt = "You are a distributed agent with Redis storage."

# 

# [agents.memory]

# backend = "redis"

# url = "redis://localhost:6379"

# prefix = "distributed_agent"

# ttl_seconds = 3600