Module core 

Modules

advisor

Configuration for the Claude Advisor server-side tool.

agent

auth

automation

commands

custom_provider

dotfile_protection

Dotfile protection configuration.

model

permissions

plugins

prompt_cache

provider

provider_override

sandbox

Sandbox configuration for VT Code

security

skills

Skills configuration

tools

Structs

AdvisorCachingConfig

Enables prompt caching for the advisor’s own transcript across calls within a conversation. This is an on/off switch, not a cache-control breakpoint.

AdvisorConfig

Configuration for the Claude Advisor server-side tool.

AgentCodexAppServerConfig

AgentConfig

Agent-wide configuration

AgentOnboardingConfig

AgentPermissionsConfig

AgentPromptSuggestionsConfig

Inline prompt suggestion configuration for the chat composer.

AnthropicConfig

Anthropic-specific provider configuration

AnthropicFallbackTarget

One explicit server-side fallback entry.

AnthropicPromptCacheSettings

Anthropic Claude cache control settings

AuthConfig

AutoPermissionConfig

Classifier-backed auto permission review configuration.

AutoPermissionEnvironmentConfig

Trust-boundary configuration for auto permission review.

AutomationConfig

Automation-specific configuration toggles.

BundledSkillsConfig

Bundled skills configuration.

CommandsConfig

Command execution configuration

CopilotAuthConfig

CustomProviderCommandAuthConfig

Command-backed bearer token configuration for a custom provider.

CustomProviderConfig

Configuration for a user-defined OpenAI-compatible provider endpoint.

CustomProviderProfileConfig

Sparse per-provider or per-model capability/profile settings.

DeepSeekPromptCacheSettings

DeepSeek prompt caching configuration (automatic KV cache reuse)

DockerSandboxConfig

Docker sandbox configuration

DotfileProtectionConfig

Dotfile protection configuration.

EditorToolConfig

External editor integration configuration

ExternalSandboxConfig

External sandbox configuration (Docker, MicroVM)

FullAutoConfig

Controls for running the agent without interactive approvals.

GatekeeperConfig

Gatekeeper mitigation configuration (macOS only)

GeminiPromptCacheSettings

Gemini API caching preferences

MemoriesConfig

Codex-compatible memories configuration.

MicroVMSandboxConfig

MicroVM sandbox configuration

ModelConfig

Model-specific behavior configuration

MoonshotPromptCacheSettings

Moonshot prompt caching configuration (leverages server-side reuse)

NetworkAllowlistEntryConfig

Network allowlist entry

NetworkConfig

Network egress configuration

OpenAIAuthConfig

OpenAIConfig

OpenAI-specific provider configuration

OpenAIHostedShellConfig

OpenAI hosted shell configuration.

OpenAIHostedShellDomainSecret

Per-domain secret injected by the OpenAI hosted shell runtime.

OpenAIHostedShellNetworkPolicy

Request-scoped network policy for OpenAI hosted shell environments.

OpenAIManualCompactionConfig

Manual compaction defaults for the native OpenAI /responses/compact flow.

OpenAIPromptCacheSettings

OpenAI prompt caching controls (automatic with metrics)

OpenAIToolSearchConfig

OpenAI hosted tool search configuration.

OpenResponsesConfig

Open Responses specification compliance configuration

OpenRouterAuthConfig

Configuration for OpenRouter OAuth authentication.

OpenRouterPromptCacheSettings

OpenRouter passthrough caching controls

PermissionsConfig

Permission system configuration - Controls command resolution, audit logging, and caching

PersistentMemoryConfig

PluginRuntimeConfig

Runtime configuration for dynamic plugin loading.

PromptCachingConfig

Global prompt caching configuration loaded from vtcode.toml

ProviderOverrideConfig

Configuration for overriding a built-in provider’s model list.

ProviderPromptCachingConfig

Per-provider configuration overrides

ResolvedCustomProviderProfile

Resolved capability/profile settings after applying provider defaults and exact model-specific overrides.

ResourceLimitsConfig

Resource limits configuration

SandboxConfig

Sandbox configuration

ScheduledTasksConfig

Controls for the built-in scheduled task subsystem.

SeccompConfig

Linux seccomp configuration

SecurityConfig

Security configuration

SensitivePathsConfig

Sensitive paths configuration

SkillsConfig

Skills system configuration

ToolSearchConfig

Configuration for Anthropic’s tool search feature (advanced-tool-use beta) Enables dynamic tool discovery for large tool catalogs (up to 10k tools)

ToolsConfig

Tools configuration

WebFetchConfig

Web Fetch tool security configuration

WebSearchConfig

Web Search tool configuration.

ZaiPromptCacheSettings

Z.AI prompt caching configuration (disabled until platform exposes metrics)

Enums

AdvisorCacheTtl

Cache lifetime for the advisor’s own transcript across calls in a conversation.

AnthropicFallbackMode

Keyword forms of AnthropicFallbacks.

AnthropicFallbacks

Server-side refusal fallback selection: "default", "off", or an explicit list of fallback models.

CustomProviderApiFormat

Typed API format used by custom providers.

ExternalSandboxType

External sandbox type

GeminiPromptCacheMode

Gemini prompt caching mode selection

MicroVmProvider

MicroVM provider (VMM) type

NetworkPolicy

Network egress policy

OpenAIHostedShellEnvironment

How VT Code should provision OpenAI hosted shell environments.

OpenAIHostedShellNetworkPolicyType

Hosted shell network access policy.

OpenAIHostedSkill

Hosted skill reference mounted into an OpenAI hosted shell environment.

OpenAIHostedSkillVersion

Hosted skill version selector for OpenAI Responses hosted shell mounts.

OpenAIPreferredMethod

OpenAIPromptCacheKeyMode

OpenAI prompt cache key derivation mode.

OpenAIServiceTier

Native OpenAI service tier selection.

PluginTrustLevel

Trust model for third-party plugins.

PromptCacheRetention

OpenAI prompt cache retention policy. Controls the model-side prompt caching window.

PromptFormat

Prompt format for skills section (Agent Skills spec)

ResourceLimitsPreset

Resource limits preset

SandboxPolicy

Sandbox policy following the Codex model

SeccompProfilePreset

Seccomp profile preset

SkillsRenderMode

Skills rendering mode

ThinkingDisplayMode

Controls how thinking content is returned in Anthropic API responses.

ToolPolicy

Tool execution policy

ToolProfile

Model-facing tool profile for a session.

ToolSearchAlgorithm

Tool search algorithm for Anthropic’s advanced-tool-use beta

WebFetchMode

Web fetch security mode

WebSearchProvider

Web search provider identifier.

Functions

build_openai_prompt_cache_key

Build a stable OpenAI prompt_cache_key for requests that should share provider-side cache routing.

build_session_affinity_prompt_cache_key

Build the namespaced session-affinity / prompt-cache key for a request.

map_prompt_cache_key_for_provider

Rewrite an OpenAI-style cache key for routes that share the wire shape but need distinct routing stickiness. Keeps the stable session identifier while namespacing gateway/session-affinity traffic apart from native OpenAI.

session_affinity_key_enabled

Whether LLMRequest.prompt_cache_key should carry session lineage for this provider.

session_affinity_provider

Providers that send lineage-stable session/cache identity on the wire.

tool_call_delay_for_rate

tool_loop_limit_reached