Crate embacle 

Embacle — LLM Runners

Standalone library providing pluggable LlmProvider implementations that delegate to CLI tools (Claude Code, Copilot, Cursor Agent, OpenCode, Gemini, Codex, Goose, Cline, Continue, Warp, Kiro, Kilo Code), an HTTP API client (OpenAI-compatible), and ACP (Copilot Headless) for LLM completions.

CLI runners wrap a binary, build prompts from ChatMessage sequences, parse JSON output, and manage session continuity. The Copilot Headless runner communicates via NDJSON-framed JSON-RPC with copilot --acp.

Two companion binary crates build on this library:

• embacle-server — OpenAI-compatible REST API + MCP Streamable HTTP on a single port
• embacle-mcp — standalone MCP server over stdio or HTTP

Progress feedback (AG-UI)

With the agui feature enabled, the agui module exposes the AG-UI protocol event vocabulary plus an emitter trait so downstream runners, agent loops, and pipelines can report run/step/tool/text progress to user-facing clients without coupling to a specific transport.

Re-exports

pub use agent::AgentExecutor;

pub use agent::AgentResult;

pub use agent::OnTurnCallback;

pub use agent::TurnInfo;

pub use auth::ProviderReadiness;

pub use cache::CacheConfig;

pub use cache::CacheProvider;

pub use cache::CacheStats;

pub use capability_guard::validate_capabilities;

pub use claude_code::ClaudeCodeRunner;

pub use cli_common::CliRunnerBase;

pub use cline_cli::ClineCliRunner;

pub use codex_cli::CodexCliRunner;

pub use compat::CliCapabilities;

pub use compat::CliFeatureFlags;

pub use config::CliRunnerType;

pub use config::RunnerConfig;

pub use container::ContainerConfig;

pub use container::ContainerExecutor;

pub use container::NetworkMode;

pub use continue_cli::ContinueCliRunner;

pub use copilot::copilot_fallback_models;

pub use copilot::CopilotRunner;

pub use cursor_agent::CursorAgentRunner;

pub use discovery::discover_runner;

pub use discovery::resolve_binary;

pub use factory::create_runner;

pub use factory::create_runner_with_config;

pub use factory::parse_runner_type;

pub use factory::valid_provider_names;

pub use factory::ALL_PROVIDERS;

pub use fallback::is_empty_completion;

pub use fallback::Attempt;

pub use fallback::FallThrough;

pub use fallback::FallbackObserver;

pub use fallback::FallbackProvider;

pub use fallback::FallthroughReason;

pub use fallback::QuotaCooldowns;

pub use fallback::ResponsePolicy;

pub use fallback::RetryConfig;

pub use fallback::Tier;

pub use fallback::DEFAULT_QUOTA_COOLDOWN;

pub use fallback::QUOTA_COOLDOWN_SKIP;

pub use gemini_cli::GeminiCliRunner;

pub use goose_cli::GooseCliRunner;

pub use guardrail::ContentLengthGuardrail;

pub use guardrail::Guardrail;

pub use guardrail::GuardrailProvider;

pub use guardrail::GuardrailViolation;

pub use guardrail::PiiScrubGuardrail;

pub use guardrail::TopicFilterGuardrail;

pub use kilo_cli::KiloCliRunner;

pub use kiro_cli::KiroCliRunner;

pub use mcp_tool_bridge::McpToolDefinition;

pub use mcp_tool_bridge::McpToolExecutor;

pub use metrics::MetricsProvider;

pub use metrics::MetricsReport;

pub use opencode::OpenCodeRunner;

pub use pricing::calculate_cost;

pub use pricing::calculate_cost_for;

pub use pricing::is_not_per_token_metered;

pub use pricing::ModelPricing;

pub use pricing::TokenCounts;

pub use pricing::NOT_PER_TOKEN_METERED_PROVIDERS;

pub use pricing::PRICING_TABLE;

pub use quality_gate::QualityGateProvider;

pub use quality_gate::QualityPolicy;

pub use structured_output::extract_json_from_response;

pub use structured_output::request_structured_output;

pub use structured_output::StructuredOutputRequest;

pub use warp_cli::WarpCliRunner;

pub use types::ImagePart;

pub use types::ResponseFormat;

pub use types::ToolCallRequest;

pub use types::ToolChoice;

pub use types::ToolDefinition;

pub use turn::ConversationTurnId;

pub use tool_simulation::execute_with_text_tools;

pub use tool_simulation::format_tool_results_as_text;

pub use tool_simulation::generate_tool_catalog;

pub use tool_simulation::inject_tool_catalog;

pub use tool_simulation::parse_tool_call_blocks;

pub use tool_simulation::strip_tool_call_blocks;

pub use tool_simulation::FunctionCall;

pub use tool_simulation::FunctionDeclaration;

pub use tool_simulation::FunctionResponse;

pub use tool_simulation::TextToolHandler;

pub use tool_simulation::TextToolResponse;

pub use agui::AgUiEmitter;

pub use agui::AgUiEvent;

pub use agui::AgUiEventFilter;

pub use agui::AgUiEventKind;

pub use agui::NoopEmitter;

pub use config_file::build_fallback_from_config;

pub use config_file::build_runner_config;

pub use config_file::load_config;

pub use config_file::load_config_from;

pub use config_file::resolve_alias;

pub use config_file::DefaultsConfig;

pub use config_file::EmbacleConfig;

pub use config_file::FallbackConfig;

pub use config_file::McpServerConfig;

pub use config_file::ProviderConfig;

pub use http_api::AnthropicConfig;

pub use http_api::AnthropicEffort;

pub use http_api::AnthropicProvider;

pub use http_api::AnthropicThinking;

pub use http_api::CohereConfig;

pub use http_api::CohereProvider;

pub use http_api::GeminiConfig;

pub use http_api::GeminiProvider;

pub use http_api::GroqConfig;

pub use http_api::GroqProvider;

pub use http_api::OpenAiCompatibleConfig;

pub use http_api::OpenAiCompatibleProvider;

pub use http_api::OpenRouterConfig;

pub use http_api::OpenRouterProvider;

pub use web_ui::WebProviderConfig;

pub use web_ui::WebUiConfig;

pub use web_ui::WebUiRunner;

pub use copilot_common::HeadlessEventStream;

pub use copilot_common::HeadlessStreamEvent;

pub use copilot_common::HeadlessToolResponse;

pub use copilot_common::HeadlessTurnProvider;

pub use copilot_common::ObservedToolCall;

pub use copilot_common::PermissionPolicy;

pub use copilot_common::DEFAULT_MAX_HISTORY_TURNS;

pub use copilot_headless::CopilotHeadlessRunner;

pub use copilot_headless_config::CopilotHeadlessConfig;

pub use copilot_sdk::CopilotSdkRunner;

pub use copilot_sdk_config::CopilotSdkConfig;

pub use copilot_sdk_config::CopilotSdkProvider;

Modules

agent

Configurable agent loop with multi-turn tool calling

agui

AG-UI protocol event schema (feature-gated).

auth

Auth readiness checking for CLI runners

cache

Response caching decorator

capability_guard

Request/provider capability validation

claude_code

Claude Code CLI runner

cli_common

Shared base struct and macro for CLI runner boilerplate

cline_cli

Cline CLI runner

codex_cli

Codex CLI runner

compat

Version compatibility and capability detection

config

Shared configuration types for CLI runners

config_file

TOML-based declarative configuration file loading

container

Container-based execution backend

continue_cli

Continue CLI runner

copilot

GitHub Copilot CLI runner

copilot_common

Types and helpers every Copilot provider shares: the permission policy, the turn API, history rendering The turn API both Copilot providers implement.

copilot_headless

Configuration for the Copilot Headless (ACP) provider

copilot_headless_config

Configuration types for the Copilot Headless provider

copilot_models

Ranked catalog of Copilot-served models for intelligent default selection and self-heal

copilot_sdk

The Copilot SDK provider — GitHub’s Rust runtime over stdio GitHub Copilot through its Rust SDK.

copilot_sdk_config

Configuration types for the Copilot SDK provider

cursor_agent

Cursor Agent CLI runner

discovery

Binary auto-detection and discovery

factory

Runner factory, provider parsing, and provider enumeration

fallback

Provider fallback chains

gemini_cli

Gemini CLI runner

goose_cli

Goose CLI runner

guardrail

Pluggable guardrail middleware for request/response validation

http_api

HTTP API providers: Anthropic, Gemini, Cohere, Groq, OpenRouter, OpenAI-compatible

kilo_cli

Kilo Code CLI runner

kiro_cli

Kiro CLI runner

mcp_tool_bridge

MCP tool definition to text-tool-simulation bridge

metrics

Cost/latency normalization decorator

opencode

OpenCode CLI runner

pool

Credential pools

pricing

Per-token price table keyed on each provider’s name()

process

Subprocess spawning with safety limits

prompt

Prompt construction from ChatMessage sequences

quality_gate

Response quality validation with retry

quota

Provider quota: snapshots, pluggable checkers, and cooldown policy. Provider quota.

quota_http

HTTP quota checkers: the real Anthropic budget, and the GitHub headroom proxy. HTTP quota checkers.

quota_store

Where routing state lives, behind a trait so a shared store is a plugin. Routing state storage.

quota_store_redis

Routing state shared through Redis across every instance spending one account. A QuotaStore that several processes share.

router

Budget-driven routing across several providers, with a pluggable strategy. Budget-driven routing.

sandbox

Environment sandboxing and tool policy

stream

Stream wrapper for child process lifecycle management

structured_output

Schema-enforced JSON output from any provider

system_delivery

Per-runner System-message delivery contract. System-message delivery contract.

tool_simulation

Text-based tool simulation for CLI runners without native function calling

turn

Conversation-turn correlation identifier threaded through a user utterance

types

Core types: traits, messages, requests, responses, and errors

warp_cli

Warp terminal oz CLI runner

web_ui

Browser-driven web-UI provider (drives Claude.ai via a headless profile)

Macros

delegate_provider_base

Generate the boilerplate LlmProvider trait methods.