Azure AI Agents Persistent SDK for .NET. Low-level SDK for creating and managing AI agents with threads, messages, runs, and tools.
Install with Tessl CLI
npx tessl i github:sickn33/antigravity-awesome-skills --skill azure-ai-agents-persistent-dotnet67
Does it follow best practices?
If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:
npx tessl skill review --optimize ./path/to/skillValidation for skill structure
Core agent polling workflow
Correct agent package
0%
100%
Azure.Identity package
100%
100%
PROJECT_ENDPOINT env var
0%
100%
MODEL_DEPLOYMENT_NAME env var
0%
100%
DefaultAzureCredential authentication
0%
100%
Administration.CreateAgentAsync
0%
100%
Threads.CreateThreadAsync
0%
100%
Messages.CreateMessageAsync
50%
100%
500ms polling delay
0%
100%
Polling status conditions
87%
100%
Ascending message order
25%
100%
MessageTextContent type check
100%
100%
Resource cleanup
100%
100%
RequestFailedException handling
100%
100%
Without context: $0.3428 · 4m 15s · 18 turns · 65 in / 6,871 out tokens
With context: $0.4205 · 3m 54s · 23 turns · 324 in / 4,308 out tokens
Function calling with tool output submission
FunctionToolDefinition used
0%
100%
BinaryData.FromObjectAsJson for parameters
0%
100%
JsonNamingPolicy.CamelCase
0%
0%
RequiresAction status detection
0%
100%
SubmitToolOutputsAction pattern
0%
100%
RequiredFunctionToolCall cast
0%
100%
SubmitToolOutputsToRunAsync call
0%
100%
toolApprovals: null parameter
0%
100%
ToolOutput construction
0%
100%
Resource cleanup
0%
100%
Without context: $1.0220 · 8m 38s · 45 turns · 99 in / 10,875 out tokens
With context: $0.4328 · 4m 23s · 21 turns · 290 in / 4,944 out tokens
File search with vector store and streaming
Files.UploadFileAsync used
75%
100%
PersistentAgentFilePurpose.Agents
50%
100%
VectorStores.CreateVectorStoreAsync
80%
100%
FileSearchToolResource with VectorStoreIds.Add
100%
100%
ToolResources FileSearch assignment
100%
100%
FileSearchToolDefinition in tools
100%
100%
CreateRunStreamingAsync used
100%
100%
MessageContentUpdate text extraction
100%
100%
StreamingUpdateReason handling
0%
100%
Full resource cleanup
100%
100%
Without context: $0.4686 · 5m 11s · 25 turns · 30 in / 7,713 out tokens
With context: $0.4172 · 4m 9s · 21 turns · 447 in / 5,536 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.