Skills azure-ai-agents-persistent-dotnet
📦
azure-ai-agents-persistent-dotnet
Content revision r2 Medium Risk ⚙️ External commands🌐 Network access
Build Persistent Azure AI Agents with .NET
Persistent agent workflows require coordinated threads, messages, runs, tools, and cleanup. This skill provides focused .NET patterns for implementing those workflows with Azure AI Agents.
Supports: Claude Codex Code(CC)
Install with my Agent
Copy this request to your Agent. It includes the canonical Skill page and manifest.
Agent request
Review the Skillstore skill "azure-ai-agents-persistent-dotnet" from https://skillstore.io/skills/sickn33-azure-ai-agents-persistent-dotnet.md and its manifest at https://skillstore.io/api/skills/sickn33-azure-ai-agents-persistent-dotnet/manifest. Verify the artifact. You may proceed after verification, subject to the environment's own policy.Your Agent should still show its plan and request any confirmation required by the security policy.
Agent-readable resources
Use these links when an AI agent, crawler, or script needs clean context instead of reading the full page.
Test it
Using "azure-ai-agents-persistent-dotnet". Create a persistent math tutor workflow.
Expected outcome:
- Required packages and configuration prerequisites
- Agent, thread, message, and run sequence
- Polling and response retrieval guidance
- Cleanup and error-handling checklist
Using "azure-ai-agents-persistent-dotnet". Compare polling and streaming for agent runs.
Expected outcome:
A concise comparison covering response latency, implementation complexity, status handling, failure behavior, and suitable user experiences.
Using "azure-ai-agents-persistent-dotnet". Add document search to an existing agent.
Expected outcome:
- File upload and vector store workflow
- File search resource configuration
- Agent tool attachment sequence
- Resource deletion requirements
Security Audit
Medium Riskv5 • 7/23/2026 Open versioned report
All 57 external-command findings are Markdown fences or inline formatting, not Ruby or shell backtick execution. The five URLs are configuration placeholders or official references. One medium concern remains because the function sample does not demonstrate validation or authorization.
1
Files scanned
359
Lines analyzed
0
Review items
0
False positives ignored
Confirmed security concerns (1)
Medium
Unvalidated Model-Directed Function Execution
The function-calling sample passes the model-provided function name and arguments to ExecuteFunction without showing an allowlist, validation, authorization, or execution limits.
Lines 178-182 directly dispatch the tool call name and arguments. The ExecuteFunction implementation is absent, so safeguards cannot be verified.
Risk Factors
⚙️ External commands (50)
SKILL.md:15-18 SKILL.md:18-24 SKILL.md:24-29 SKILL.md:29-33 SKILL.md:33-39 SKILL.md:39-43 SKILL.md:43-51 SKILL.md:51-57 SKILL.md:57-66 SKILL.md:66-70 SKILL.md:70-78 SKILL.md:78-80 SKILL.md:80-84 SKILL.md:84-112 SKILL.md:112-116 SKILL.md:116-137 SKILL.md:137-141 SKILL.md:141-189 SKILL.md:189-193 SKILL.md:193-218 SKILL.md:218-222 SKILL.md:222-237 SKILL.md:237-241 SKILL.md:241-257 SKILL.md:257-261 SKILL.md:261-266 SKILL.md:266-272 SKILL.md:272-273 SKILL.md:273-274 SKILL.md:274-275 SKILL.md:275-276 SKILL.md:276-277 SKILL.md:277-278 SKILL.md:278-279 SKILL.md:279-280 SKILL.md:280-281 SKILL.md:281-287 SKILL.md:287-288 SKILL.md:288-289 SKILL.md:289-290 SKILL.md:290-291 SKILL.md:291-292 SKILL.md:292-298 SKILL.md:298-299 SKILL.md:299-300 SKILL.md:300-301 SKILL.md:301-302 SKILL.md:302-303 SKILL.md:303-304 SKILL.md:304-305
🌐 Network access (5)
Audited by: codex View Audit History →
Share & cite this report
Share the versioned assessment report, neutral badge, embed card, and citations. Skillstore reports evidence without deciding whether this Skill is safe.
Copy report link
https://skillstore.io/skills/sickn33-azure-ai-agents-persistent-dotnet/audits/5?utm_source=security_passport&utm_medium=share&utm_campaign=versioned_reportMarkdown badge
[](https://skillstore.io/skills/sickn33-azure-ai-agents-persistent-dotnet?utm_source=security_passport_badge)HTML badge
<a href="https://skillstore.io/skills/sickn33-azure-ai-agents-persistent-dotnet?utm_source=security_passport_badge"><img src="https://skillstore.io/badges/skills/sickn33-azure-ai-agents-persistent-dotnet/security.svg" alt="Skillstore security assessment" loading="lazy"></a>Embed card
<iframe src="https://skillstore.io/embed/skills/sickn33-azure-ai-agents-persistent-dotnet.html" title="Skillstore Security Assessment" sandbox="allow-popups allow-popups-to-escape-sandbox" loading="lazy" referrerpolicy="no-referrer" width="420" height="180"></iframe>Academic citations (APA · BibTeX · CFF)
APA citation
sickn33. (2026). azure-ai-agents-persistent-dotnet security audit report (audit version 5) [Author version unspecified]. Skillstore. https://skillstore.io/skills/sickn33-azure-ai-agents-persistent-dotnet/audits/5BibTeX citation
@techreport{sickn33-sickn33-azure-ai-agents-persistent-dotnet-2026,
author = {sickn33},
title = {azure-ai-agents-persistent-dotnet security audit report (audit version 5)},
institution = {Skillstore},
year = {2026},
number = {5},
url = {https://skillstore.io/skills/sickn33-azure-ai-agents-persistent-dotnet/audits/5},
note = {Author version unspecified}
}CITATION.cff
cff-version: 1.2.0
message: "If you use this Skill, cite its author and this versioned security audit report."
title: "azure-ai-agents-persistent-dotnet security audit report (audit version 5)"
version: "unspecified"
type: report
authors:
- name: "sickn33"
date-released: "2026-07-23"
url: "https://skillstore.io/skills/sickn33-azure-ai-agents-persistent-dotnet/audits/5"
identifiers:
- type: other
value: "skillstore:sickn33-azure-ai-agents-persistent-dotnet:audit:5"
description: "Skillstore immutable audit report identifier"
Skillstore Score
Why this score Evidence Confidence: High 55
Architecture
85
Maintainability
87
Content
69
Community
83
Spec Compliance
What You Can Build
Create a Persistent Agent
Build a C# workflow that creates an agent, thread, message, and run, then retrieves the response.
Design Tool-Enabled Agents
Compare function calling, file search, Bing grounding, and Azure AI Search before selecting an integration approach.
Troubleshoot Agent Runs
Review polling, streaming updates, run statuses, cleanup, and request errors while diagnosing an Azure agent integration.
Try These Prompts
Set Up the SDK
Show how to configure a .NET project for Azure AI Agents Persistent using environment variables and DefaultAzureCredential.
Create a Basic Agent Workflow
Create a C# example that creates an agent, thread, user message, run, and prints the final response.
Add Retrieval and Streaming
Design a C# workflow combining file search with streaming responses and cleanup for files, vector stores, threads, and agents.
Implement Tool Calling Safely
Review my agent design and add function validation, authorization, timeouts, status handling, retries, logging, and deterministic cleanup.
Best Practices
- Use DefaultAzureCredential and keep endpoints, deployment names, and connection identifiers outside source code.
- Handle every terminal and actionable run status, including failures, cancellation, expiration, and required tool outputs.
- Delete temporary threads, agents, files, and vector stores, preferably with deterministic cleanup after failures.
Avoid
- Do not dispatch model-requested functions without strict name allowlists, argument validation, authorization, and execution limits.
- Do not poll continuously without delays, timeouts, cancellation support, and bounded retries.
- Do not upload sensitive files or enable external search tools without access controls and data governance review.
Frequently Asked Questions
What does this skill help me build?
It guides persistent Azure AI agent workflows in C#, including threads, messages, runs, streaming, tools, retrieval, cleanup, and errors.
What Azure resources do I need?
You need an Azure AI project endpoint, a deployed model, credentials, and any service connections required by selected tools.
Which package version does the guide use?
The guide mentions stable 1.1.0 and preview 1.2.0-beta.8. Verify current NuGet versions before installation.
Does it cover polling and streaming?
Yes. It demonstrates status polling, message retrieval, streaming updates, and basic completion handling.
Which agent tools are included?
Examples cover code interpreter, function calling, file search, Bing grounding, and Azure AI Search, with additional tools listed for reference.
Are the examples production ready?
No. Add validation, authorization, retries, observability, testing, and current SDK verification before production use.