Class: Ace::LLM::Providers::CLI::OpenCodeClient

Inherits:
Organisms::BaseClient
  • Object
show all
Includes:
CliArgsSupport
Defined in:
lib/ace/llm/providers/cli/open_code_client.rb

Overview

Client for interacting with OpenCode CLI Provides access to multiple AI providers through OpenCode’s unified platform

Constant Summary collapse

API_BASE_URL =

Not used for CLI interaction but required by BaseClient

"https://models.dev"
DEFAULT_GENERATION_CONFIG = 
{}.freeze
DEFAULT_MODEL =

Default model (can be overridden by config)

"google/gemini-2.5-flash"

Class Method Summary collapse

Instance Method Summary collapse

Constructor Details

#initialize(model: nil, **options) ⇒ OpenCodeClient

Returns a new instance of OpenCodeClient.



31
32
33
34
35
36
 
# File 'lib/ace/llm/providers/cli/open_code_client.rb', line 31

def initialize(model: nil, **options)
  @model = model || DEFAULT_MODEL
  # Skip normal BaseClient initialization that requires API key
  @options = options
  @generation_config = options[:generation_config] || {}
end

Class Method Details

.provider_name ⇒ Object

Provider registration - auto-registers as “opencode”



24
25
26
 
# File 'lib/ace/llm/providers/cli/open_code_client.rb', line 24

def self.provider_name
  "opencode"
end

Instance Method Details

#generate(messages, **options) ⇒ Hash

Generate a response from the LLM

Parameters:

  • messages (Array<Hash>) —

    Conversation messages

  • options (Hash) —

    Generation options

Returns:

  • (Hash) —

    Response with text and metadata



47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
 
# File 'lib/ace/llm/providers/cli/open_code_client.rb', line 47

def generate(messages, **options)
  validate_opencode_availability!

  # Convert messages to prompt format
  prompt = format_messages_as_prompt(messages)

  # Build full prompt with system instruction for accurate token accounting
  full_prompt = build_full_prompt(prompt, options)

  cmd = build_opencode_command_with_prompt(full_prompt, options)
  stdout, stderr, status = execute_opencode_command(cmd, options: options)

  parse_opencode_response(stdout, stderr, status, full_prompt, options)
rescue => e
  handle_opencode_error(e)
end

#list_models ⇒ Object

List available OpenCode models



65
66
67
68
69
70
71
72
73
74
75
76
77
 
# File 'lib/ace/llm/providers/cli/open_code_client.rb', line 65

def list_models
  # Return a standard set of models that OpenCode typically supports
  # Actual models come from YAML config
  [
    {id: "google/gemini-2.5-flash", name: "Gemini 2.5 Flash", description: "Fast Google model", context_size: 1_000_000},
    {id: "google/gemini-2.0-flash-experimental", name: "Gemini 2.0 Flash", description: "Experimental Google model", context_size: 1_000_000},
    {id: "google/gemini-1.5-pro", name: "Gemini 1.5 Pro", description: "Advanced Google model", context_size: 2_000_000},
    {id: "anthropic/claude-3-5-sonnet", name: "Claude 3.5 Sonnet", description: "Anthropic model", context_size: 200_000},
    {id: "anthropic/claude-3-5-haiku", name: "Claude 3.5 Haiku", description: "Fast Anthropic model", context_size: 200_000},
    {id: "openai/gpt-4o", name: "GPT-4 Omni", description: "OpenAI model", context_size: 128_000},
    {id: "openai/gpt-4o-mini", name: "GPT-4 Omni Mini", description: "Small OpenAI model", context_size: 128_000}
  ]
end

#needs_credentials? ⇒ Boolean

Override to indicate this client doesn’t need API credentials

Returns:

  • (Boolean) 


39
40
41
 
# File 'lib/ace/llm/providers/cli/open_code_client.rb', line 39

def needs_credentials?
  false
end