.NET: added Mem0Sharp integration for in-memory storage in agent samples.

[Jay Khawaja (jihadkhawaja)]

#7467

[Jay Khawaja (jihadkhawaja)]

Jay Khawaja (Jay Khawaja (@jihadkhawaja)) please read the following Contributor License Agreement(CLA). If you agree with the CLA, please reply with the following information.

@microsoft-github-policy-service agree [company="{your company}"]

Options:

• (default - no company specified) I have sole ownership of intellectual property rights to my Submissions and I am not making Submissions in the course of work for my employer.

@microsoft-github-policy-service agree

• (when company given) I am making Submissions in the course of work for my employer (or my employer has intellectual property rights in my Submissions by contract or applicable law). I have permission from my employer to make Submissions and enter into this Agreement on behalf of my employer. By signing below, the defined term “You” includes me and my employer.

@microsoft-github-policy-service agree company="Microsoft"

Contributor License Agreement

@microsoft-github-policy-service agree

[Copilot]

Pull request overview

Adds a Mem0Sharp-based in-memory provider sample demonstrating memory sharing across agent sessions.

Changes:

• Adds the Mem0Sharp provider and runnable sample.
• Documents configuration and persistence alternatives.
• Registers the dependency and sample project.

Reviewed changes

Copilot reviewed 7 out of 7 changed files in this pull request and generated 4 comments.

Show a summary per file

File	Description
AgentWithMemory/README.md	Links the new sample.
Step09.../README.md	Documents setup and storage providers.
Step09.../Program.cs	Demonstrates cross-session recall.
Step09.../Mem0SharpProvider.cs	Implements memory storage and retrieval.
Step09...csproj	Defines dependencies and framework reference.
Directory.Packages.props	Pins Mem0Sharp 0.2.1.
agent-framework-dotnet.slnx	Registers the sample project.

Suppressed comments (2)

dotnet/samples/02-agents/AgentWithMemory/AgentWithMemory_Step09_MemoryUsingMem0Sharp/Mem0SharpProvider.cs:28

• GetAllAsync injects every memory for the user on every invocation. With the persistent stores documented for this provider, prompt size and retrieval cost grow without bound and unrelated memories are labeled as relevant; use the current external request text to call Mem0Sharp's bounded SearchAsync instead.

        var memories = await memory.GetAllAsync(
            new MemoryFilter(UserId: userId),
            cancellationToken: cancellationToken);

dotnet/samples/02-agents/AgentWithMemory/AgentWithMemory_Step09_MemoryUsingMem0Sharp/Mem0SharpProvider.cs:34

• These memories originate verbatim from user messages, but placing them in Instructions promotes that untrusted text to the model's instruction channel. A user can persist an instruction in one session and have it override behavior in later sessions. Return the recalled text as a user-role context message instead, matching the repository's built-in memory providers.

                Instructions = $"Relevant memories for this user:\n{string.Join(Environment.NewLine, memories.Select(item => $"- {item.Text}"))}",

---

💡 Add a code-review agent skill for context-aware, tailored reviews. Learn more in the docs.

[Roger Barreto (rogerbarreto)]

Hi Jay Khawaja (@jihadkhawaja) Thanks for the contribution.

Please make this change as a README.md only file that points to your MAF sample in the original repo.

https://github.com/jihadkhawaja/mem0sharp/blob/main/samples/AgentFrameworkMemory/README.md

Additionally would be interesting to make your library also multi TFM net8,9 + netstandard 2.0 friendly for a more broader audience.

[Jay Khawaja (jihadkhawaja)]

Roger Barreto (@rogerbarreto) done, Also added support for multi TFM (net8,9 and netstandard 2.0), thanks!