Azure AI Agents Persistent SDK for Java. Low-level SDK for creating and managing AI agents with threads, messages, runs, and tools.
Install with Tessl CLI
npx tessl i github:boisenoise/skills-collections --skill azure-ai-agents-persistent-java73
Quality
60%
Does it follow best practices?
Impact
100%
1.63xAverage score across 3 eval scenarios
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/antigravity-azure-ai-agents-persistent-java/SKILL.mdCore agent conversation workflow
Correct groupId
100%
100%
Correct artifactId
0%
100%
Correct version
25%
100%
Endpoint from env var
50%
100%
PersistentAgentsClientBuilder usage
25%
100%
DefaultAzureCredential auth
100%
100%
createAgent call
62%
100%
createThread call
62%
100%
createMessage with USER role
62%
100%
createRun call
37%
100%
Polling loop with QUEUED/IN_PROGRESS
0%
100%
500ms polling delay
0%
100%
listMessages retrieval
62%
100%
Resource cleanup
71%
100%
Without context: $0.4852 · 4m 56s · 21 turns · 28 in / 8,193 out tokens
With context: $0.4277 · 3m 35s · 22 turns · 184 in / 5,403 out tokens
Run status handling and error resilience
Correct Maven artifact
0%
100%
HttpResponseException import
100%
100%
Try/catch for agent creation
100%
100%
Status code in error output
100%
100%
RequiresAction status check
100%
100%
Failed status check
100%
100%
Cancelled status check
100%
100%
Polling loop present
100%
100%
Thread cleanup
100%
100%
Agent cleanup
100%
100%
Without context: $0.5422 · 4m 54s · 24 turns · 77 in / 9,718 out tokens
With context: $0.4853 · 4m 10s · 23 turns · 28 in / 7,132 out tokens
Async client for high-concurrency workloads
Correct Maven artifact
0%
100%
Correct version
0%
100%
Async client used
0%
100%
Async client builder
0%
100%
Endpoint from env var
0%
100%
DefaultAzureCredential auth
100%
100%
Concurrent execution
100%
100%
Thread cleanup
100%
100%
Agent cleanup
100%
100%
Without context: $0.4875 · 4m 39s · 18 turns · 22 in / 9,154 out tokens
With context: $0.5887 · 3m 59s · 22 turns · 27 in / 10,489 out tokens
Table of Contents
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.