Skip to main content

Crate axllm

Crate axllm 

Source

Re-exports§

pub use mcp::AxMCPClient;
pub use mcp::AxMCPOAuthOptions;
pub use mcp::AxMCPStdioTransport;
pub use mcp::AxMCPStreamableHTTPTransport;
pub use mcp::AxMCPTokenSet;
pub use mcp::AxMCPTransport;

Modules§

mcp

Structs§

AxACE
Agentic Context Engineering optimizer (Generator -> Reflector -> Curator).
AxAgent
AxBalancer
AxBootstrapFewShot
AxError
AxFlow
AxGEPA
AxGen
AxMemory
AxPlaybook
A live, evolving context playbook bound to a program. Mirrors the TypeScript AxPlaybook: grow it offline from examples (evolve), keep it growing online from live feedback (update), render it into the program context (apply_to), and persist/restore it (to_json/load). The evolution engine (ACE) is hidden behind this surface, just as optimize() hides GEPA.
AxSignature
Field
FieldType
MultiServiceRouter
OpenAICompatibleClient
ProcessCodeRuntime
ProcessCodeSession
ProviderRouter
RuntimeCapabilities
RuntimeEnvelope
ScriptedRealtimeTransport
ScriptedTransport
SignatureBuilder
Tool
ToolBuilder

Enums§

RealtimeTransport
Transport seam for the realtime turn driver: ScriptedRealtimeTransport for deterministic offline tests, the websocket-backed variant (behind the realtime feature) for live turns.

Traits§

AxAIClient
AxCodeRuntime
AxCodeSession
AxProgram
AxTransport
OptimizerEngine

Functions§

agent
agent_with_options
agent_with_search_callbacks
Construct an agent with native host search callbacks for memories and skills. The callbacks run host-side when the actor calls recall()/discover(); passing them at construction auto-enables the memory + skill subsystems (so the actor’s prompt advertises recall()/discover()), mirroring the TS/Python API. Callbacks take and return JSON values: memories (searches, alreadyLoaded) -> results, skills (searches) -> results.
ai
ax
f
flow
get_supported_ai_models
optimize
parse_json
playbook
Build an evolving context AxPlaybook for a program. Pass the student model under "studentAI" (and optionally "teacherAI"); the bound program runs with the student while reflection/curation use the teacher. When teacher is None the student is reused for reflection/curation.
run_conformance_fixture
runtime_protocol_fixture_server_main
s
signature_from_spec
stable_stringify
tool

Type Aliases§

AIClient
AnthropicClient
AxAIService
AxAgentClarificationError
AxBaseAI
AxHostCallable
A host callable the agent runtime can expose to actor code (e.g. the built-in llmQuery). JSON-typed both ways so it crosses the runtime boundary cleanly; Send + Sync so a concrete runtime may store it.
AxResult
AzureOpenAIClient
CohereClient
DeepSeekClient
GoogleGeminiClient
GrokClient
MistralClient
OpenAIResponsesClient
OptimizedArtifact
OptimizerEvaluator
RekaClient
RuntimeProtocolClient
RuntimeTransport