Loading…
Topic
12 posts about Agent Development Kit. Every summary links to the original.
See the idea
A2A lets agents discover each other and hand off tasks over JSON-RPC 2.0. This guide walks through exposing an Agent Card, sending typed messages, connecting a remote service, and routing failures to a human review checkpoint.
The Good Stuff
Paid picks we use ourselves. Links carry a referral code — a subscription through one earns us a commission at no extra cost to you.
Newest first
Google tested Gemini 4 Argon on google/adk-python, a 205,000-line Python framework with 800 source files and 17,000 unit tests. The model ran unattended for 11 hours, guided by a /goal prompt and .agents/skills rulefiles that encoded maintainer conventions. Argon first wrote a benchmark script, spent 70 minutes measuring before editing code, and proposed six optimizations across 32 files; two shipped after human code review. The shipped changes cut in-memory session prep time by about 40% in long sessions and sped up cold package imports by 21% by filtering events before copying and using lazy module accessors. The other four proposals were rejected for relying on fragile state, private library internals, or synthetic-only gains, illustrating a division of labor where the agent explores and the maintainer subtracts.
AI agents in production often pass health checks while silently failing at real tasks, such as confirming a seat without checking availability. Google and ADK released AQuA (Ambient Quality Agent), a sidecar system that continuously sweeps production agent trajectories from Cloud Trace, Cloud Logging, or BigQuery outside the request path. AQuA samples up to 1,000 recent sessions, grades each against a developer-defined goal and a nine-point breakdown checklist using a single-pass judge (or optional Gemini AutoRaters), clusters failures by mechanism, verifies clusters with a second model, and tracks them in BigQuery. Developers can trigger root-cause analysis that reads failing trajectories alongside the immutable deploy-time source snapshot, citing exact code lines without auto-applying edits. The system runs unattended, sweeps after each deploy or on demand, and produces verified failure clusters to feed back into the inner development loop, solving quality degradation that standard health checks miss.
Agent Anomaly Detection is a reasoning-based oversight and audit layer for autonomous agents on the Gemini Enterprise Agent Platform, now in Private Preview. It analyzes existing reasoning traces, tool calls, execution flow, logs, and OpenTelemetry traces asynchronously, using a lightweight statistical pass followed by deeper LLM-based examination of flagged sessions. Its initial detectors cover tool misuse, identity and privilege abuse, cascading failures, rogue agents, resource exhaustion, and token usage escalation, with findings assigned severity, probability, explanations, and recommended fixes and surfaced in Security Command Center. In the Inventory Agent example, repeated large-batch, offset-jumping list_inventory calls are classified as systematic scraping and produce a Critical resource-exhaustion finding at 95% probability; an API can let ADK callbacks or plugins block later actions. Custom business logic detectors are described as coming soon.
Part 2 of the Zero-trust Agents series addresses attacks that deterministic, build-time checks cannot recognize when syntax remains valid. It keeps the Customer Support and Returns Agent, which reads orders, calculates restocking fees in Agent Sandbox, and commits payouts through an issue_refund tool signed with a hardware-backed Cloud KMS key, then moves enforcement to Gemini Enterprise Agent Platform. Agent Gateway applies three runtime controls: Model Armor screens prompts and responses, Semantic Governance Policies evaluates proposed tool calls against intent, history, and business rules, and Agent Anomaly Detection analyzes behavior across a session. The examples cover prompt injection, a non-refundable digital-license refund, and a multi-turn refund drain; detected anomalies can trigger a new policy that blocks subsequent calls without redeploying the agent.
ADK for Kotlin 1.0 is announced as a general availability release for building AI agents with Kotlin across Android, JVM server applications, and Java-integrated systems. The release targets a production-ready toolkit while preserving an idiomatic, lightweight, composable approach, and it reaches feature parity with ADK 1.0 Core, including hierarchical agents, context compaction, human-in-the-loop confirmation, resumable sessions, annotated tools, and Java interoperability. A Kotlin Multiplatform core stays agnostic to model backends, session providers, and memory systems, while Android extensions support LiteRT-LM, ML Kit (beta), Firebase AI Logic, Room, AppSearch, and file artifacts. The database incident example combines compile-time KSP-generated function schemas, suspend-capable @Tool functions, and dynamically loaded SKILL.md playbooks with progressive disclosure. The release provides 1.0.0 dependencies, documentation, a repository, and sample applications.
The strongest submissions in the Google for Startups AI Agents Challenge repeatedly used four engineering patterns rather than merely labeling prompt chains as multi-agent systems. They exposed internally used tools through bidirectional MCP, used typed events and asyncio queues to let independent agents run concurrently, and routed fallback outputs through one shared validation function. They also reduced inference cost with tiered routing: regex or a cheap classifier handled simple intent before requests reached an expensive reasoning model, with one team reporting more than 40 percent handled before a real model call. The post argues these practices improve bounded data access, latency, resilience, and cost without requiring larger teams or newer models, and notes that Agent Development Kit and Agents CLI submissions showed them frequently.
ADK now supports native evaluation of live, voice-based agents, addressing regressions in tool use, context continuity, turn-taking, timing, and recovery that may not appear in demos. The walkthrough builds a graph-based workflow of three single-purpose agents running gemini-live-2.5-flash-native-audio, with a mocked date-of-birth tool between stages and session state carried across the open audio stream. It shows how to author scenario-based or fixed conversations, synthesize simulated user turns with Gemini TTS, configure live mode and rubric-based multi-turn trajectory judges, and run evaluations from the CLI or AgentEvaluator. ADK Web then reconstructs the audio as transcripts with playable clips, while configurable voices, languages, invocation limits, thresholds, and per-turn metrics support repeatable debugging and CI/CD regression checks.
Connecting an autonomous Agent Development Kit (ADK) agent to databases, internal APIs, and code execution turns model decisions into production-state changes that system prompts cannot securely constrain. The post presents an open-source Customer Support & Returns Agent built with ADK and Gemini, using an example where prompt injection could produce an unauthorized refund or leak environment secrets. Its zero-trust design combines hardware-backed signing for database mutations, gVisor isolation with no network egress and resource limits for generated Python, and deterministic semantic gateways for prompts, tool calls, and database updates. Local equivalents use HMAC keys, a database ingress verifier, a constrained Docker runner, and Python validation tests, while production mapping uses Cloud KMS, HSM, and VPC Service Controls. The conclusion is that infrastructure-enforced boundaries let the model reason dynamically within strict limits.
At Sonoma Raceway, Google Developer Experts built an AI-powered race coach to turn vehicle telemetry into split-second driving advice, extending an earlier pilot that handled basic telemetry. The system uses Antigravity for stateful orchestration and ingestion, Python and Jetpack Compose for mobile processing and display, Agent Development Kit for agent coordination, and hybrid Gemma 4 edge reasoning plus Gemini API cloud analysis. A custom USB interface connected a Pixel 10 directly to the vehicle telemetry network, providing a 10 Hz stream from hundreds of sensors, while an activated on-device TPU reached 40 tokens per second for local coaching alerts delivered through text-to-speech. The test identified a throttle application zone in Turn 2 associated with a 0.1-second gain, and the initiative is scheduled to continue at Interlagos, Brazil.
ADK 2.0 addresses the structural challenge of moving AI agents from prototypes into production, where infinite loops, hallucinated bypasses, and failures without clean exceptions make model-led orchestration unreliable. Because language models are slow, costly, and variable for routing, scheduling, and error handling, the release separates execution routing from language processing and reserves LLMs for ambiguous or cognitive steps. Building on ADK v1’s support for Python, Java, Go, TypeScript, and Kotlin, it adds a structured workflow runtime and task-collaboration model, with Workflows available in Python since March and newly launched for Go. Deterministic graphs combine tool calls, Human-in-the-Loop steps, and specialized agents with explicit branching and error handling. Dynamic workflows extend this model with native Python and asyncio control flows for retries, branching, and reusable sub-workflows.
ADK for Go 2.0 introduces a graph-based workflow engine for composing multi-agent applications whose execution can classify, branch, fan out, gather results, pause for human input, retry, and loop. Graphs consist of typed nodes and routed edges, with constructors for functions, emitting functions, agents, tools, joins, parallel workers, state-bound nodes, dynamic orchestration, and nested workflows; a scheduler handles concurrency, persistence, backpressure, cancellation, and resumption. Its built-in human-in-the-loop primitive validates responses, supports handoff or re-entry, and can resume durably after process restarts, while agent modes and a unified node runtime align single agents with full graphs. The release is highly additive for ADK 1.0 users, though context and event APIs require mechanical migrations, and it provides workflow examples for basic, routing, dynamic, HITL, and complex graphs.
The post presents a contract-compliance pipeline in which a Python agent uses Gemini to extract contract terms and a Go service applies deterministic corporate-policy checks. It connects the services with the Agent2Agent (A2A) protocol and uses Google's Agent Development Kit (ADK), including RemoteA2aAgent, to expose the remote validator as a local sub-agent in a SequentialAgent. A2A supplies discovery through Agent Cards at /.well-known/agent.json, JSON-RPC 2.0 messaging through message/send, typed Message Parts, and task states such as submitted, working, completed, and failed. Pipeline state is shared through ADK's ToolContext.state, with failures transitioning to MANUAL_REVIEW for human legal review when the Go service is unavailable. The accompanying open-source application demonstrates live handoffs, policy verdicts, and simulated network faults, while the conclusion emphasizes language independence, decoupling, resilience, and deterministic auditing.
The best engineering reads, once a week.
