Configuration

Deep dive into sekuire.yml - the single source of truth for your agent.

---

File Structure

sekuire.yml

project:
  name: "my-agent"
  version: "1.0.0"
  description: "A helpful AI assistant"

agents:
  assistant:
    name: "AI Assistant"
    system_prompt: "./prompts/assistant.md"
    tools: "./tools.json"
    llm:
      provider: "openai"
      model: "gpt-4-turbo"
      api_key_env: "OPENAI_API_KEY"
      temperature: 0.7
      max_tokens: 2000
    memory:
      type: "buffer"
      max_messages: 20

deployment:
  docker_image: "myrepo/my-agent:latest"
  runtime: "node"
  port: 8000

---

Project Section

Field	Type	Description
name	string	Agent project name (used in registry)
version	string	Semantic version (e.g., 1.0.0)
description	string?	Optional description

---

Agents Section

Define one or more agents. Each agent has its own LLM config, tools, and memory.

Agent Fields

Field	Type	Description
name	string	Display name
system_prompt	string	Path to system prompt markdown file
tools	string | array	Path to tools.json or inline tool list
llm	object	LLM provider configuration
memory	object?	Conversation memory settings

LLM Configuration

Field	Type	Description
provider	string	openai, anthropic, google, ollama
model	string	Model name (e.g., gpt-4-turbo)
api_key_env	string	Environment variable for API key
temperature	number?	Sampling temperature (0-2)
max_tokens	number?	Max tokens in response
base_url	string?	Custom API base URL

Memory Configuration

Field	Type	Description
type	string	buffer (FIFO) or window (sliding)
max_messages	number	Max messages to retain

---

Deployment Section

Field	Type	Description
docker_image	string	Container image reference
runtime	string	node, python, rust
port	number	Container port (default: 8000)

---

Multi-Agent Configuration

Define multiple agents in the same project:

sekuire.yml

agents:
  researcher:
    name: "Research Agent"
    system_prompt: "./prompts/researcher.md"
    llm:
      provider: "anthropic"
      model: "claude-3-opus-20240229"
      api_key_env: "ANTHROPIC_API_KEY"

  coder:
    name: "Coding Agent"
    system_prompt: "./prompts/coder.md"
    llm:
      provider: "openai"
      model: "gpt-4-turbo"
      api_key_env: "OPENAI_API_KEY"
    tools:
      - calculator
      - file_read
      - file_write

---

Environment Variables

Reference environment variables with api_key_env:

.env

OPENAI_API_KEY=sk-...
ANTHROPIC_API_KEY=sk-ant-...
GOOGLE_API_KEY=AIza...

warning

Never hardcode API keys in sekuire.yml. Always use environment variables.

---

Next Steps

• Deployment - Production options
• CLI Init - Scaffold with templates