ace-llm 0.30.0

checksums.yaml

@@ -0,0 +1,7 @@
+---
+SHA256:
+  metadata.gz: 91f463c56920b3bdaf3db5b85526ee15e8b15f101bb027e2acaf17537b6980e6
+  data.tar.gz: 56287d9afefbaed517b9d0a450c2d9c31a3e5b6b05c752e3f94097e029615466
+SHA512:
+  metadata.gz: 3d4706d5e1120f324a5f02d85766ba5dd8f49506366ee92e85f25434348bcad680826a22f7abf0802cc8bb4e0feed12acb0692e3d68d8289418d4649a1b80804
+  data.tar.gz: 23a2fb71311c4c95713fa5f6e03c65fd0aec10dba13c2703f81fbec6f397c757bd61aee24b2cc06f565ebaef0e2f1db927779219944639dfb853820888157731

data/.ace-defaults/llm/config.yml

@@ -0,0 +1,31 @@
+# ace-llm default configuration
+# Override in .ace/llm/config.yml for project-specific settings
+
+llm:
+  # Optional allow-list of active providers.
+  # When unset, null, or empty, all discovered providers remain available.
+  # Example:
+  # providers:
+  #   active:
+  #     - google
+  #     - anthropic
+  providers: {}
+
+  # Default timeout in seconds for LLM requests
+  # CLI providers (opencode, claude, codex, gemini) may need longer
+  # due to subprocess overhead and "thinking" phases
+  timeout: 120
+
+  # Global fallback policy shared by QueryInterface and ace-llm CLI query.
+  fallback:
+    enabled: true
+    retry_count: 3
+    retry_delay: 1.0
+    max_total_timeout: 30.0
+    chains: {}
+    providers: []
+
+  # Default context limit for unknown models (in tokens)
+  # Provider-specific limits are defined in providers/*.yml
+  context_limit:
+    default: 128000

data/.ace-defaults/llm/presets/claude/prompt.yml

@@ -0,0 +1,5 @@
+timeout: 600
+# Prompt-only mode: Claude receives only the composed prompt payload.
+cli_args:
+  - --tools
+  - ""

data/.ace-defaults/llm/presets/claude/ro.yml

@@ -0,0 +1,6 @@
+timeout: 600
+# Keep @ro tool-assisted for richer reviews while constraining Claude to
+# prompt inspection and shell access only.
+cli_args:
+  - --tools
+  - Bash,Read

data/.ace-defaults/llm/presets/claude/rw.yml

@@ -0,0 +1,4 @@
+timeout: 600
+cli_args:
+  - --permission-mode
+  - acceptEdits

data/.ace-defaults/llm/presets/claude/yolo.yml

@@ -0,0 +1,3 @@
+timeout: 600
+cli_args:
+  - dangerously-skip-permissions

data/.ace-defaults/llm/presets/codex/ro.yml

@@ -0,0 +1,5 @@
+timeout: 600
+cli_args:
+  - --full-auto
+  - --sandbox
+  - read-only

data/.ace-defaults/llm/presets/codex/rw.yml

@@ -0,0 +1,3 @@
+timeout: 600
+cli_args:
+  - --full-auto

data/.ace-defaults/llm/presets/codex/yolo.yml

@@ -0,0 +1,3 @@
+timeout: 600
+cli_args:
+  - --dangerously-bypass-approvals-and-sandbox

data/.ace-defaults/llm/presets/gemini/ro.yml

@@ -0,0 +1,4 @@
+timeout: 600
+cli_args:
+  - --approval-mode
+  - plan

data/.ace-defaults/llm/presets/gemini/rw.yml

@@ -0,0 +1,4 @@
+timeout: 600
+cli_args:
+  - --approval-mode
+  - auto_edit

data/.ace-defaults/llm/presets/gemini/yolo.yml

@@ -0,0 +1,4 @@
+timeout: 600
+cli_args:
+  - --approval-mode
+  - yolo

data/.ace-defaults/llm/presets/opencode/ro.yml

@@ -0,0 +1 @@
+timeout: 600

data/.ace-defaults/llm/presets/opencode/rw.yml

@@ -0,0 +1 @@
+timeout: 600

data/.ace-defaults/llm/presets/opencode/yolo.yml

@@ -0,0 +1,3 @@
+timeout: 600
+cli_args:
+  - dangerously-skip-permissions

data/.ace-defaults/llm/presets/pi/ro.yml

@@ -0,0 +1 @@
+timeout: 600

data/.ace-defaults/llm/presets/pi/rw.yml

@@ -0,0 +1 @@
+timeout: 600

data/.ace-defaults/llm/presets/pi/yolo.yml

@@ -0,0 +1 @@
+timeout: 600

data/.ace-defaults/llm/providers/anthropic.yml

@@ -0,0 +1,34 @@
+name: anthropic
+last_synced: 2026-03-23
+class: Ace::LLM::Organisms::AnthropicClient
+gem: ace-llm
+context_limit: 1000000  # Claude models have 1M context window
+models:
+  - claude-haiku-4-5
+  - claude-haiku-4-5-20251001
+  - claude-opus-4-5
+  - claude-opus-4-5-20251101
+  - claude-sonnet-4-5
+  - claude-sonnet-4-5-20250929
+aliases:
+  global:
+    opus: anthropic:o
+    sonnet: anthropic:s
+    haiku: anthropic:h
+    csonnet: anthropic:s
+  model:
+    s: claude-sonnet-4-5
+    o: claude-opus-4-5
+    h: claude-haiku-4-5
+api_key:
+  env: ANTHROPIC_API_KEY
+  required: true
+  description: Anthropic API key
+capabilities:
+  - text_generation
+  - streaming
+  - function_calling
+  - vision
+default_options:
+  temperature: 0.7
+  max_tokens: 16384

data/.ace-defaults/llm/providers/google.yml

@@ -0,0 +1,36 @@
+name: google
+last_synced: 2026-02-27
+class: Ace::LLM::Organisms::GoogleClient
+gem: ace-llm
+context_limit: 1000000  # Gemini models have 1M+ context window
+models:
+  - gemini-2.5-flash
+  - gemini-2.5-flash-lite
+  - gemini-2.5-pro
+  - gemini-3.1-pro-preview
+  - gemini-3-flash-preview
+  - gemini-flash-latest
+  - gemini-flash-lite-latest
+aliases:
+  global:
+    gflash: google:flash
+    gpro: google:pro
+    glite: google:lite
+  model:
+    flash: gemini-flash-latest
+    lite: gemini-flash-lite-latest
+    pro: gemini-2.5-pro
+    pro-preview: gemini-3.1-pro-preview
+    flash-preview: gemini-3-flash-preview
+api_key:
+  env: GEMINI_API_KEY
+  required: true
+  description: Google AI Studio API key
+capabilities:
+  - text_generation
+  - streaming
+  - function_calling
+  - vision
+default_options:
+  temperature: 0.7
+  max_tokens: 16384

data/.ace-defaults/llm/providers/groq.yml

@@ -0,0 +1,29 @@
+name: groq
+class: Ace::LLM::Organisms::GroqClient
+gem: ace-llm
+context_limit: 128000  # Groq models typically have 128K context window
+models:
+  - openai/gpt-oss-120b
+  - openai/gpt-oss-20b
+  - moonshotai/kimi-k2-instruct-0905
+aliases:
+  global:
+    groq: groq:openai/gpt-oss-120b
+    groq-fast: groq:openai/gpt-oss-20b
+    groq-kimi: groq:moonshotai/kimi-k2-instruct-0905
+    groq-saba: groq:mistral-saba-24b
+  model:
+    gpt-oss: openai/gpt-oss-120b
+    gpt-oss-120b: openai/gpt-oss-120b
+    gpt-oss-20b: openai/gpt-oss-20b
+    kimi-k2: moonshotai/kimi-k2-instruct-0905
+    saba: mistral-saba-24b
+api_key:
+  env: GROQ_API_KEY
+  required: true
+  description: Groq API key
+capabilities:
+  - text_generation
+default_options:
+  temperature: 0.7
+  max_tokens: 16384

data/.ace-defaults/llm/providers/lmstudio.yml

@@ -0,0 +1,24 @@
+name: lmstudio
+last_synced: 2025-12-05
+class: Ace::LLM::Organisms::LMStudioClient
+gem: ace-llm
+context_limit: 128000  # Default context limit; varies by local model
+# Models are user-defined locally in LM Studio and not synced from models.dev.
+# This list is intentionally empty.
+models:
+aliases:
+  global:
+    local: lmstudio:local
+  model:
+    local: local-model
+    custom: custom-model
+endpoint: http://localhost:1234
+api_key:
+  required: false
+  description: LM Studio local server (no API key needed)
+capabilities:
+  - text_generation
+  - streaming
+default_options:
+  temperature: 0.7
+  max_tokens: 16384

data/.ace-defaults/llm/providers/mistral.yml

@@ -0,0 +1,33 @@
+name: mistral
+last_synced: 2025-12-05
+class: Ace::LLM::Organisms::MistralClient
+gem: ace-llm
+context_limit: 128000  # Mistral Large has 128K context window
+models:
+  - codestral-latest
+  - devstral-medium-2507
+  - devstral-small-2507
+  - mistral-large-latest
+  - mistral-medium-2508
+  - mistral-medium-latest
+  - mistral-small-latest
+aliases:
+  global:
+    mistral-large: mistral:large
+    mistral-medium: mistral:medium
+  model:
+    large: mistral-large-latest
+    medium: mistral-medium-latest
+    small: mistral-small-latest
+    code: codestral-latest
+api_key:
+  env: MISTRAL_API_KEY
+  required: true
+  description: Mistral AI API key
+capabilities:
+  - text_generation
+  - streaming
+  - code_generation
+default_options:
+  temperature: 0.7
+  max_tokens: 16384

data/.ace-defaults/llm/providers/openai.yml

@@ -0,0 +1,33 @@
+name: openai
+last_synced: 2025-12-05
+class: Ace::LLM::Organisms::OpenAIClient
+gem: ace-llm
+context_limit: 1050000  # GPT-5.x has 1.05M context window
+models:
+  - gpt-5.1
+  - gpt-5.1-chat-latest
+  - gpt-5.1-codex
+  - gpt-5.1-codex-max
+  - gpt-5.1-codex-mini
+aliases:
+  global:
+    oagpt: openai:gpt
+    oacodex: openai:codex
+    oamax: openai:max
+  model:
+    gpt: gpt-5.1
+    codex: gpt-5.1-codex
+    max: gpt-5.1-codex-max
+api_key:
+  env: OPENAI_API_KEY
+  required: true
+  description: OpenAI API key
+capabilities:
+  - text_generation
+  - streaming
+  - function_calling
+  - vision
+  - embeddings
+default_options:
+  temperature: 0.7
+  max_tokens: 16384

data/.ace-defaults/llm/providers/openrouter.yml

@@ -0,0 +1,45 @@
+name: openrouter
+class: Ace::LLM::Organisms::OpenRouterClient
+gem: ace-llm
+context_limit: 128000  # Default context limit; varies by model
+models:
+  # Fast inference (:nitro = throughput priority → Groq/Cerebras)
+  - openai/gpt-oss-120b:nitro
+  - openai/gpt-oss-20b:nitro
+  - moonshotai/kimi-k2-0905:nitro
+  - qwen/qwen3-max
+  # DeepSeek (exclusive to OpenRouter)
+  - deepseek/deepseek-v3.2
+  # Kimi/Moonshot
+  - moonshotai/kimi-k2-thinking
+  # Qwen
+  - qwen/qwen3-coder
+  # Others
+  - z-ai/glm-4.6
+  - minimax/minimax-m1
+aliases:
+  models:
+    # Fast inference (nitro = throughput)
+    oss-nitro: openrouter:openai/gpt-oss-120b:nitro
+    ss-small-nitro: openrouter:openai/gpt-oss-20b:nitro
+    kimi-nitro: openrouter:moonshotai/kimi-k2-0905:nitro
+    qwen3: openrouter:qwen/qwen3-max
+    # Kimi
+    kimi: openrouter:moonshotai/kimi-k2-0905
+    kimi-think: openrouter:moonshotai/kimi-k2-thinking
+    # DeepSeek
+    deepseek: openrouter:deepseek/deepseek-v3.2
+    # Qwen
+    qwen-coder: openrouter:qwen/qwen3-coder
+    # Others
+    glm: openrouter:z-ai/glm-4.6
+    minimax: openrouter:minimax/minimax-m1
+api_key:
+  env: OPENROUTER_API_KEY
+  required: true
+  description: OpenRouter API key
+capabilities:
+  - text_generation
+default_options:
+  temperature: 0.7
+  max_tokens: 16384

data/.ace-defaults/llm/providers/togetherai.yml

@@ -0,0 +1,26 @@
+name: togetherai
+last_synced: 2025-12-05
+class: Ace::LLM::Organisms::TogetherAIClient
+gem: ace-llm
+context_limit: 128000  # Default context limit; varies by model
+models:
+  - Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8
+  - deepseek-ai/DeepSeek-V3
+  - moonshotai/Kimi-K2-Instruct
+  - openai/gpt-oss-120b
+aliases:
+  model:
+    qwen: Qwen/Qwen3-Coder-480B-A35B-Instruct-FP8
+    deepseek: deepseek-ai/DeepSeek-V3
+    kimi: moonshotai/Kimi-K2-Instruct
+    oss: openai/gpt-oss-120b
+api_key:
+  env: TOGETHER_API_KEY
+  required: true
+  description: Together AI API key
+capabilities:
+  - text_generation
+  - streaming
+default_options:
+  temperature: 0.7
+  max_tokens: 16384

data/.ace-defaults/llm/providers/xai.yml

@@ -0,0 +1,30 @@
+name: xai
+last_synced: 2025-12-06
+class: Ace::LLM::Organisms::XAIClient
+gem: ace-llm
+context_limit: 131072  # Grok models have 131K context window
+models:
+  - grok-4
+  - grok-4-1-fast
+  - grok-4-1-fast-non-reasoning
+  - grok-4-fast-non-reasoning
+  - grok-code-fast-1
+aliases:
+  global:
+    grok: xai:grok-4
+    grokfast: xai:grok-4-1-fast
+    grokcode: xai:grok-code-fast-1
+  model:
+    grok: grok-4
+    fast: grok-4-1-fast
+    fast-non: grok-4-1-fast-non-reasoning
+    code: grok-code-fast-1
+api_key:
+  env: XAI_API_KEY
+  required: true
+  description: x.ai API key
+capabilities:
+  - text_generation
+default_options:
+  temperature: 0.7
+  max_tokens: 16384

data/.ace-defaults/llm/providers/zai.yml

@@ -0,0 +1,18 @@
+name: zai
+last_synced: 2026-02-27
+class: Ace::LLM::Organisms::ZaiClient
+gem: ace-llm
+context_limit: 128000
+models:
+  - glm-4.7-flashx
+  - glm-4.7
+  - glm-5
+api_key:
+  env: ZAI_API_KEY
+  required: true
+  description: Z.AI API key
+capabilities:
+  - text_generation
+default_options:
+  temperature: 0.7
+  max_tokens: 16384

data/.ace-defaults/llm/thinking/claude/high.yml

@@ -0,0 +1,3 @@
+cli_args:
+  - --effort
+  - high

data/.ace-defaults/llm/thinking/claude/low.yml

@@ -0,0 +1,3 @@
+cli_args:
+  - --effort
+  - low

data/.ace-defaults/llm/thinking/claude/medium.yml

@@ -0,0 +1,3 @@
+cli_args:
+  - --effort
+  - medium

data/.ace-defaults/llm/thinking/claude/xhigh.yml

@@ -0,0 +1,3 @@
+cli_args:
+  - --effort
+  - high

data/.ace-defaults/llm/thinking/codex/high.yml

@@ -0,0 +1,3 @@
+cli_args:
+  - -c
+  - model_reasoning_effort="high"

data/.ace-defaults/llm/thinking/codex/low.yml

@@ -0,0 +1,3 @@
+cli_args:
+  - -c
+  - model_reasoning_effort="low"

data/.ace-defaults/llm/thinking/codex/medium.yml

@@ -0,0 +1,3 @@
+cli_args:
+  - -c
+  - model_reasoning_effort="medium"

data/.ace-defaults/llm/thinking/codex/xhigh.yml

@@ -0,0 +1,3 @@
+cli_args:
+  - -c
+  - model_reasoning_effort="high"

data/.ace-defaults/nav/protocols/guide-sources/ace-llm.yml

@@ -0,0 +1,10 @@
+---
+# Guide Sources Protocol Configuration for ace-llm gem
+name: ace-llm
+type: gem
+description: Guides from ace-llm gem
+priority: 10
+config:
+  relative_path: handbook/guides
+  pattern: "*.g.md"
+  enabled: true