Serialization logo

Serialization

CommunityPopular
Aaronontheweb
serialization

Choose the right serialization format for .NET applications. Prefer schema-based formats (Protobuf, MessagePack) over reflection-based (Newtonsoft.Json). Use System.Text.Json with AOT source generators for JSON scenarios.

Overview

PublisherAaronontheweb
Repositorydotnet-skills
Skill nameserialization
Stars
1.2K
Forks
110
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by Aaronontheweb on GitHub. Read the source before you install it.

Installation

Install the Serialization AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/Aaronontheweb/dotnet-skills/tree/master/skills/serialization
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/Aaronontheweb/dotnet-skills.git /tmp/dotnet-skills
mkdir -p .claude/skills
cp -r /tmp/dotnet-skills/skills/serialization .claude/skills/serialization
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Serialization in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Serialization on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Serialization is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Serialization in .NET

When to Use This Skill

Use this skill when:

  • Choosing a serialization format for APIs, messaging, or persistence
  • Migrating from Newtonsoft.Json to System.Text.Json
  • Implementing AOT-compatible serialization
  • Designing wire formats for distributed systems
  • Optimizing serialization performance

Schema-Based vs Reflection-Based

AspectSchema-BasedReflection-Based
ExamplesProtobuf, MessagePack, System.Text.Json (source gen)Newtonsoft.Json, BinaryFormatter
Type info in payloadNo (external schema)Yes (type names embedded)
VersioningExplicit field numbers/namesImplicit (type structure)
PerformanceFast (no reflection)Slower (runtime reflection)
AOT compatibleYesNo
Wire compatibilityExcellentPoor

Recommendation: Use schema-based serialization for anything that crosses process boundaries.


Format Recommendations

Use CaseRecommended FormatWhy
REST APIsSystem.Text.Json (source gen)Standard, AOT-compatible
gRPCProtocol BuffersNative format, excellent versioning
Actor messagingMessagePack or ProtobufCompact, fast, version-safe
Event sourcingProtobuf or MessagePackMust handle old events forever
CachingMessagePackCompact, fast
ConfigurationJSON (System.Text.Json)Human-readable
LoggingJSON (System.Text.Json)Structured, parseable

Formats to Avoid

FormatProblem
BinaryFormatterSecurity vulnerabilities, deprecated, never use
Newtonsoft.Json defaultType names in payload break on rename
DataContractSerializerComplex, poor versioning
XMLVerbose, slow, complex

System.Text.Json with Source Generators

For JSON serialization, use System.Text.Json with source generators for AOT compatibility and performance.

Setup

csharp
// Define a JsonSerializerContext with all your types
[JsonSerializable(typeof(Order))]
[JsonSerializable(typeof(OrderItem))]
[JsonSerializable(typeof(Customer))]
[JsonSerializable(typeof(List<Order>))]
[JsonSourceGenerationOptions(
    PropertyNamingPolicy = JsonKnownNamingPolicy.CamelCase,
    DefaultIgnoreCondition = JsonIgnoreCondition.WhenWritingNull)]
public partial class AppJsonContext : JsonSerializerContext { }

Usage

csharp
// Serialize with context
var json = JsonSerializer.Serialize(order, AppJsonContext.Default.Order);

// Deserialize with context
var order = JsonSerializer.Deserialize(json, AppJsonContext.Default.Order);

// Configure in ASP.NET Core
builder.Services.ConfigureHttpJsonOptions(options =>
{
    options.SerializerOptions.TypeInfoResolverChain.Insert(0, AppJsonContext.Default);
});

Benefits

  • No reflection at runtime - All type info generated at compile time
  • AOT compatible - Works with Native AOT publishing
  • Faster - No runtime type analysis
  • Trim-safe - Linker knows exactly what's needed

Protocol Buffers (Protobuf)

Best for: Actor systems, gRPC, event sourcing, any long-lived wire format.

Setup

bash
dotnet add package Google.Protobuf
dotnet add package Grpc.Tools

Define Schema

protobuf
// orders.proto
syntax = "proto3";

message Order {
    string id = 1;
    string customer_id = 2;
    repeated OrderItem items = 3;
    int64 created_at_ticks = 4;

    // Adding new fields is always safe
    string notes = 5;  // Added in v2 - old readers ignore it
}

message OrderItem {
    string product_id = 1;
    int32 quantity = 2;
    int64 price_cents = 3;
}

Versioning Rules

protobuf
// SAFE: Add new fields with new numbers
message Order {
    string id = 1;
    string customer_id = 2;
    string shipping_address = 5;  // NEW - safe
}

// SAFE: Remove fields (old readers ignore unknown, new readers use default)
// Just stop using the field, keep the number reserved
message Order {
    string id = 1;
    // customer_id removed, but field 2 is reserved
    reserved 2;
}

// UNSAFE: Change field types
message Order {
    int32 id = 1;  // Was: string - BREAKS!
}

// UNSAFE: Reuse field numbers
message Order {
    reserved 2;
    string new_field = 2;  // Reusing 2 - BREAKS!
}

MessagePack

Best for: High-performance scenarios, compact payloads, actor messaging.

Setup

bash
dotnet add package MessagePack
dotnet add package MessagePack.Annotations

Usage with Contracts

csharp
[MessagePackObject]
public sealed class Order
{
    [Key(0)]
    public required string Id { get; init; }

    [Key(1)]
    public required string CustomerId { get; init; }

    [Key(2)]
    public required IReadOnlyList<OrderItem> Items { get; init; }

    [Key(3)]
    public required DateTimeOffset CreatedAt { get; init; }

    // New field - old readers skip unknown keys
    [Key(4)]
    public string? Notes { get; init; }
}

// Serialize
var bytes = MessagePackSerializer.Serialize(order);

// Deserialize
var order = MessagePackSerializer.Deserialize<Order>(bytes);

AOT-Compatible Setup

csharp
// Use source generator for AOT
[MessagePackObject]
public partial class Order { }  // partial enables source gen

// Configure resolver
var options = MessagePackSerializerOptions.Standard
    .WithResolver(CompositeResolver.Create(
        GeneratedResolver.Instance,  // Generated
        StandardResolver.Instance));

Migrating from Newtonsoft.Json

Common Issues

NewtonsoftSystem.Text.JsonFix
$type in JSONNot supported by defaultUse discriminators or custom converters
JsonPropertyJsonPropertyNameDifferent attribute
DefaultValueHandlingDefaultIgnoreConditionDifferent API
NullValueHandlingDefaultIgnoreConditionDifferent API
Private settersRequires [JsonInclude]Explicit opt-in
Polymorphism[JsonDerivedType] (.NET 7+)Explicit discriminators

Migration Pattern

csharp
// Newtonsoft (reflection-based)
public class Order
{
    [JsonProperty("order_id")]
    public string Id { get; set; }

    [JsonProperty(NullValueHandling = NullValueHandling.Ignore)]
    public string? Notes { get; set; }
}

// System.Text.Json (source-gen compatible)
public sealed record Order(
    [property: JsonPropertyName("order_id")]
    string Id,

    string? Notes  // Null handling via JsonSerializerOptions
);

[JsonSerializable(typeof(Order))]
[JsonSourceGenerationOptions(
    PropertyNamingPolicy = JsonKnownNamingPolicy.SnakeCaseLower,
    DefaultIgnoreCondition = JsonIgnoreCondition.WhenWritingNull)]
public partial class OrderJsonContext : JsonSerializerContext { }

Polymorphism with Discriminators

csharp
// .NET 7+ polymorphism
[JsonDerivedType(typeof(CreditCardPayment), "credit_card")]
[JsonDerivedType(typeof(BankTransferPayment), "bank_transfer")]
public abstract record Payment(decimal Amount);

public sealed record CreditCardPayment(decimal Amount, string Last4) : Payment(Amount);
public sealed record BankTransferPayment(decimal Amount, string AccountNumber) : Payment(Amount);

// Serializes as:
// { "$type": "credit_card", "amount": 100, "last4": "1234" }

Wire Compatibility Patterns

Tolerant Reader

Old code must safely ignore unknown fields:

csharp
// Protobuf/MessagePack: Automatic - unknown fields skipped
// System.Text.Json: Configure to allow
var options = new JsonSerializerOptions
{
    UnmappedMemberHandling = JsonUnmappedMemberHandling.Skip
};

Introduce Read Before Write

Deploy deserializers before serializers for new formats:

csharp
// Phase 1: Add deserializer (deployed everywhere)
public Order Deserialize(byte[] data, string manifest) => manifest switch
{
    "Order.V1" => DeserializeV1(data),
    "Order.V2" => DeserializeV2(data),  // NEW - can read V2
    _ => throw new NotSupportedException()
};

// Phase 2: Enable serializer (next release, after V1 deployed everywhere)
public (byte[] data, string manifest) Serialize(Order order) =>
    _useV2Format
        ? (SerializeV2(order), "Order.V2")
        : (SerializeV1(order), "Order.V1");

Never Embed Type Names

csharp
// BAD: Type name in payload - renaming class breaks wire format
{
    "$type": "MyApp.Order, MyApp.Core",
    "id": "123"
}

// GOOD: Explicit discriminator - refactoring safe
{
    "type": "order",
    "id": "123"
}

Performance Comparison

Approximate throughput (higher is better):

FormatSerializeDeserializeSize
MessagePack★★★★★★★★★★★★★★★
Protobuf★★★★★★★★★★★★★★★
System.Text.Json (source gen)★★★★☆★★★★☆★★★☆☆
System.Text.Json (reflection)★★★☆☆★★★☆☆★★★☆☆
Newtonsoft.Json★★☆☆☆★★☆☆☆★★★☆☆

For hot paths, prefer MessagePack or Protobuf.


Akka.NET Serialization

For Akka.NET actor systems, use schema-based serialization:

hocon
akka {
  actor {
    serializers {
      messagepack = "Akka.Serialization.MessagePackSerializer, Akka.Serialization.MessagePack"
    }
    serialization-bindings {
      "MyApp.Messages.IMessage, MyApp" = messagepack
    }
  }
}

See Akka.NET Serialization Docs.


Best Practices

DO

csharp
// Use source generators for System.Text.Json
[JsonSerializable(typeof(Order))]
public partial class AppJsonContext : JsonSerializerContext { }

// Use explicit field numbers/keys
[MessagePackObject]
public class Order
{
    [Key(0)] public string Id { get; init; }
}

// Use records for immutable message types
public sealed record OrderCreated(OrderId Id, CustomerId CustomerId);

DON'T

csharp
// Don't use BinaryFormatter (ever)
var formatter = new BinaryFormatter();  // Security risk!

// Don't embed type names in wire format
settings.TypeNameHandling = TypeNameHandling.All;  // Breaks on rename!

// Don't use reflection serialization for hot paths
JsonConvert.SerializeObject(order);  // Slow, not AOT-compatible

Resources

Frequently asked questions

What does the Serialization AI skill do?

Choose the right serialization format for .NET applications. Prefer schema-based formats (Protobuf, MessagePack) over reflection-based (Newtonsoft.Json). Use System.Text.Json with AOT source generators for JSON scenarios.

Why use Serialization on TypingMind?

Because you install it once and use it with any model. Serialization is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Serialization in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Aaronontheweb/dotnet-skills/tree/master/skills/serialization. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use Serialization?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Serialization?

As many as you like. As long as a model supports skills, you can use Serialization with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Serialization AI skill free?

Yes. It is published on GitHub by Aaronontheweb under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇