Skip to Content
These docs describe donkey-kit v0.2.1. Release notes
Feature overview

Live

Feature overview

DDK is gateway-aware: control stays at the proxy, efficiency moves into the agent. The gateway enforces policy; DDK makes each decision visible and actionable in your code, so the agent reacts to a refusal, paces its budget and reports what it did.

Gateway-aware: control at the proxy, efficiency in the agent. The agent sends a request through DDK to the gateway; 403 and 429 responses come back through DDK to the agent. The DDK capabilities: budget, refusals, simulate, conformance, OpenTelemetry and native framework objects.

Everything below hangs off a single shared transport inside your process — the one place where every request enters and every response leaves. Each capability attaches there once, so you never wire it call by call.

Model access

FrameworkCallReturns
LangGraphdonkey.langgraph.chat_model("gpt-4o")langchain_openai.ChatOpenAI
Google ADKdonkey.adk.model("gpt-4o")LiteLlm
Google ADK on a Format=Gemini proxydonkey.adk.gemini("gemini-3.8-flash")google.adk.models.Gemini
Strandsdonkey.strands.model("gpt-4o")OpenAIModel
MS Agent Frameworkdonkey.agent_framework.chat_client("gpt-4o")Agent Framework chat client
LlamaIndexdonkey.llamaindex.llm("gpt-4o")OpenAILike
OpenAI Agents SDKdonkey.openai_agents.model("gpt-4o")OpenAIChatCompletionsModel
Anthropic SDKdonkey.anthropic.client()anthropic.AsyncAnthropic
CrewAIdonkey.crewai.llm("gpt-4o")crewai.BaseLLM

Every adapter offers the same governed connection three ways — a factory on a shared Donkey (donkey.langgraph.chat_model(...)), a module-level factory (from donkey_kit.integrations.langgraph import chat_model), or connection_kwargs() when you want to build the native object yourself.

Governance

!Typed refusals

Purpose: turn every gateway rejection into a typed exception — PIIDetected, TokenBudgetExceeded, RequestRateLimitExceeded, PromptInjectionBlocked, ContentSafetyBlocked, AuthError, GatewayUnavailable and more. Goal: branch on the governance outcome, not on a parsed error body. Advantage: a PII block is never mistaken for an auth failure, and a policy 429 is never retried.

Live
↺Refusal handlers

Purpose: register a reaction to a typed refusal once with donkey.on(...): wait for the reset, mask and resend, fall back to another model, or escalate. Goal: one PII handler instead of one per graph node. Advantage: applied at the transport and bounded, and no handler can turn a refusal into a success.

Live
◷Budget & pacing

Purpose: expose the gateway’s token window as a Budget object. Goal: stop before the limit, not after it. Advantage: pace(reserve=) and wait_for_reset() let an overnight batch slow down instead of dying at 2am.

Live
◐Identity

Purpose: on-behalf-of user identity. Goal: per-user policy reaches the gateway. Advantage: requests never silently fall back to the service identity.

Live
⏸Human-in-the-loop

Purpose: one vocabulary for “pause and ask a human”. Goal: @donkey.tool(approval="required") pauses the call until someone decides. Advantage: the approver lands on the span next to the correlation ID. Client-side and advisory; it never overrides a refusal.

Live
⚖Policy handshake

Purpose: read the policy set in force. Advantage: skip calls that are certain to be refused. Advisory only — the gateway always wins.

Roadmap

Observability

Developer tooling

Registry & catalog

What DDK leaves to the platform

DDK makes the platform’s capabilities reachable and typed; it does not reproduce them. Policy enforcement, semantic caching, provisioning, agent scanners, kill switch, trusted agent identity, approval UIs and evaluation all stay with Agent Fabric and Omni Gateway. Where the platform exposes a signal, such as a semantic-cache hit or the routing decision, DDK surfaces it to your code. See the Roadmap for the full list.

Last updated on