Add Garnet Command

microsoft/add-garnet-command

Adds a new built-in RESP command to Garnet end-to-end. Covers enum registration, parsing, dispatch, RESP handler, API surface, storage session, RMW callbacks, command metadata JSON, ACL tests, and integration tests. Use when asked to "add a command", "implement RI.SET", "add RESP command", or any new server command. Do NOT use for custom extension commands (CustomRawStringFunctions) or object-type sub-operations.

10k tokens

context cost

the whole folder, loaded on every use

files

instructions only

copies elsewhere

how many repositories repackaged it

11916

stars on the repo

on the repository, not the skill itself

Install

one command, takes just this skill from the repository

npx skills add https://github.com/microsoft/garnet --skill add-garnet-command

What it tells the agent to use

found in the instruction text

Read reads your files

Write writes files

The instruction itself

41 sections, as written by the author

Add a New Built-in RESP Command to Garnet

Step-by-step guide for implementing a new built-in RESP command in Garnet. This covers every file that must be created or modified, the tools that must be run, caveats discovered during implementation, and how to verify correctness.

Scope: Built-in commands that are part of the Garnet server (not custom extensions registered via REGISTERCS).

Overview: What Must Change

Adding a single new command touches at minimum these areas:

|---|------|-------|-----------|

| 4 | RESP handler | libs/server/Resp/<Category>/*.cs | ✅ Always |

| 7 | Storage session | libs/server/Storage/Session/[Main\|Object\|Unified]Store/*Ops.cs | If key-value command (not blocking/admin) |

Step 1: Add the RespCommand Enum Value

File: libs/server/Resp/Parser/RespCommand.cs

The RespCommand enum is writes-first: write (mutating) commands occupy a dense,

explicitly-numbered block immediately after NONE, followed by reads, scripts, and non-data:

Write commands:    APPEND = 1 ... BITOP_DIFF = 120   (EXPLICIT values; the ONLY persisted values)
Read commands:     BITCOUNT ... RISCAN               (auto-numbered; never persisted)
Script commands:   EVAL, EVALSHA                     (LastDataCommand = EVALSHA)
Non-key commands:  PING, SUBSCRIBE, etc.
Admin commands:    AUTH, CONFIG, etc.

Why writes-first: the write command's numeric value is persisted to disk (serialized into

RespInputHeader.cmd in the AOF and streamed to replicas). Keeping writes in a dense,

explicitly-numbered leading block makes those values stable — adding a read/admin command

(which is never persisted) cannot shift them.

Read/write classification uses enum ranges (positional constants, not hardcoded values):

RespCommandExtensions.FirstWriteCommand (APPEND=1) .. LastWriteCommand (BITOP_DIFF=120) → write
FirstReadCommand .. LastReadCommand (= EVAL - 1) → read-only

Rules for adding a command:

Write (mutating) command: APPEND it at the end of the write block, immediately before

the // Read-only commands marker, with the next explicit value (e.g. MYCMD = 121,).

Then update LastWriteCommand if it is no longer BITOP_DIFF. Never insert into the middle

of the write block or change/reuse an existing value — persisted values must be append-only.

Read-only / script / non-data command: add it in the appropriate (auto-numbered) section.

These are never persisted, so their values may shift freely.

Adding a write shifts the auto-numbered reads up by one — that is harmless (reads aren't

persisted) and requires no other change.

Guardrails (do not bypass):

PersistedEnumStabilityTests snapshots the write-block values and asserts the write range

contains exactly the write commands. If you change an existing write value it fails — that is a

breaking on-disk change requiring an AofHeader.AofHeaderVersion bump + a legacy remap in

AofProcessor / LegacyRespCommand, not an edited expected value.

If a new write command's value ever approaches the custom command range (0xFEFF), that is also

caught by a test.

Boundary markers to watch (search for these comments):

APPEND = 1, // Note: FirstWriteCommand — keep as the first write command
BITOP_DIFF = 120, // Note: LastWriteCommand — append new write commands after this with the next value
EVALSHA, // Note: Update LastDataCommand if adding new data commands after this

Step 2: Add Command Parsing

File: libs/server/Resp/Parser/RespCommand.cs

Two parsing tiers exist:

Hash table path: `RespCommandHashLookup` (primary path for most commands)

The hash table in libs/server/Resp/Parser/RespCommandHashLookupData.cs provides O(1) lookup for all built-in commands. This is the recommended path for all new commands.

To add a new primary command, add one line to PopulatePrimaryTable() in RespCommandHashLookupData.cs:

Add("DELIFGREATER", RespCommand.DELIFGREATER);

To add a command with subcommands (e.g., MYPARENT SUBCMD):

Add the parent with hasSub: true in PopulatePrimaryTable():

   Add("MYPARENT", RespCommand.MYPARENT, hasSub: true);

Define the subcommand array in RespCommandHashLookupData.cs:

   private static readonly (string Name, RespCommand Command)[] MyparentSubcommands =
   [
       ("SUBCMD1", RespCommand.MYPARENT_SUBCMD1),
       ("SUBCMD2", RespCommand.MYPARENT_SUBCMD2),
   ];

Build the table in the static constructor in RespCommandHashLookup.cs:

   myparentSubTable = BuildSubTable(MyparentSubcommands, out myparentSubTableMask);

Wire it into LookupSubcommand() in RespCommandHashLookup.cs:

   RespCommand.MYPARENT => (myparentSubTable, myparentSubTableMask),

⚠️ Important: Use the exact wire-protocol spelling for the hash table name string. Some commands use hyphens (e.g., "SET-CONFIG-EPOCH" not "SETCONFIGEPOCH"). Check CmdStrings.cs for the canonical spelling.

⚠️ Convention: Define the command name string in libs/server/Resp/CmdStrings.cs and reference it from the parser, rather than using inline "..."u8 literals. This keeps command name strings centralized and reusable (e.g., for error messages).

// In CmdStrings.cs:
public static ReadOnlySpan<byte> DELIFGREATER => "DELIFGREATER"u8;

SIMD fast path: `FastParseCommand()` (optional, for hottest commands only)

Static Vector128<byte> patterns defined in libs/server/Resp/Parser/RespCommandSimdPatterns.cs that match the full RESP encoding (*N\r\n$L\r\nCMD\r\n) in a single 16-byte comparison. Use the RespPattern(argCount, "CMD") helper to create new patterns. Only needed for the most performance-critical commands with:

Fixed argument count (single digit)
Command names of 3-6 characters (total encoded length must fit in 16 bytes)

Most new commands should NOT be added here — the hash table + MRU cache provide excellent performance for all commands. Only add SIMD patterns if benchmarking shows the command is a bottleneck.

Step 3: Add Command Dispatch

File: libs/server/Resp/RespServerSession.cs

Three dispatch methods exist, and which one you use matters for latency tracking:

| Method | For | Latency |

|--------|-----|---------|

| ProcessBasicCommands | Fast single/dual-arg commands | @fast only |

| ProcessArrayCommands | Fast multi-arg commands | @fast only |

| ProcessOtherCommands | Slow commands, admin commands | @slow OK |

⚠️ WARNING: Do NOT add @slow-classified commands to ProcessBasicCommands or ProcessArrayCommands. This breaks latency tracking. If in doubt, use ProcessOtherCommands.

Pattern:

RespCommand.MYCMD => NetworkMYCMD(ref storageApi),

Add before the _ => ... fallthrough in the appropriate method.

Step 4: Implement the RESP Handler

New file or existing file in libs/server/Resp/

Command handlers are methods on the RespServerSession partial class:

private bool NetworkMYCMD<TGarnetApi>(ref TGarnetApi storageApi)
    where TGarnetApi : IGarnetApi
{
    // 1. Validate argument count
    if (parseState.Count != N)
        return AbortWithWrongNumberOfArguments(nameof(RespCommand.MYCMD));

    // 2. Validate other inputs (short-circuit before going to storage)
    var key = parseState.GetArgSliceByRef(0);
    // e.g., parse and validate optional flags, numeric arguments, etc.
    if (!parseState.TryGetInt(1, out var _))
    {
        WriteError(CmdStrings.RESP_ERR_GENERIC_VALUE_IS_NOT_INTEGER);
        return true;
    }

    // 3. Build input/output and call storage API
    // Note: To avoid double-parsing a parameter, you can pass a pre-parsed
    // value in the input struct's auxiliary arguments (e.g., input.arg1).
    var input = new StringInput(RespCommand.MYCMD, ref parseState, startIdx: 1);
    var output = GetStringOutput();
    var status = storageApi.MyOperation(key, ref input, ref output);

    // 4. Write RESP response
    if (status == GarnetStatus.OK)
    {
        ProcessOutput(output);
    }
    else
    {
        WriteError(CmdStrings.RESP_ERR_MY_MESSAGE);
    }

    return true;
}

Key patterns:

Arguments: parseState.GetArgSliceByRef(i) returns ref PinnedSpanByte
Input/Output: Instantiate StringInput/StringOutput (for string commands), ObjectInput/ObjectOutput (for object commands), or UnifiedInput/UnifiedOutput (for unified commands) before calling the storage API
Response (happy path): Use ProcessOutput(output) in the common case — this handles writing the RESP response from the output struct
Response (errors/special cases): Use RespServerSession extension methods (e.g., WriteError(...), WriteDirect(...), WriteInt64(...), etc.) — these handle SendAndReset() internally
Error strings: Store as u8 literals in CmdStrings (e.g., CmdStrings.RESP_ERR_MY_MESSAGE) rather than inline
Always return true — there are no partial executions

Object command RESP handler pattern

Object commands (Hash, List, Set, SortedSet) follow a similar pattern to string commands. The main difference is that the RESP handler uses ObjectInput/ObjectOutput with the appropriate operation enum and must handle WRONGTYPE errors:

File: libs/server/Resp/Objects/[ObjectName]Commands.cs (e.g., SortedSetCommands.cs)

private unsafe bool SortedSetAdd<TGarnetApi>(ref TGarnetApi storageApi)
    where TGarnetApi : IGarnetApi
{
    if (parseState.Count < 3)
        return AbortWithWrongNumberOfArguments("ZADD");

    var key = parseState.GetArgSliceByRef(0);

    var header = new RespInputHeader(GarnetObjectType.SortedSet) { SortedSetOp = SortedSetOperation.ZADD };
    var input = new ObjectInput(header, ref parseState, startIdx: 1);
    var output = GetObjectOutput();

    var status = storageApi.SortedSetAdd(key, ref input, ref output);

    switch (status)
    {
        case GarnetStatus.WRONGTYPE:
            while (!RespWriteUtils.TryWriteError(CmdStrings.RESP_ERR_WRONG_TYPE, ref dcurr, dend))
                SendAndReset();
            break;
        default:
            ProcessOutput(output.SpanByteAndMemory);
            break;
    }

    return true;
}

Key differences from string commands:

Uses ObjectInput with a RespInputHeader(GarnetObjectType.XXX) and an operation enum (e.g., SortedSetOperation.ZADD)
Must handle GarnetStatus.WRONGTYPE — object commands can fail if the key holds a different object type
The actual data operation logic lives in libs/server/Objects/[ObjectName]/[ObjectName]ObjectImpl.cs, dispatched via the operation enum

For new object sub-operations:

Add a value to the [ObjectName]Operation enum in libs/server/Objects/[ObjectName]/[ObjectName]Object.cs

and handle it in the Operate method's switch statement.

> ⚠️ Persisted, append-only: the object operation enums (HashOperation, ListOperation,

> SetOperation, SortedSetOperation) have explicit values and are serialized to the AOF via

> RespInputHeader.SubId. Append the new operation at the end with the next explicit value;

> never insert into the middle, reorder, or change an existing value. PersistedEnumStabilityTests

> guards these values. The id occupies a full header byte, so up to 256 sub-operations per type are

> allowed.

Unified command note

Unified commands (EXISTS, DELETE, TYPE, TTL, EXPIRE, RENAME, etc.) are type-agnostic — they work on both raw string and object values. The RESP handler pattern is the same as string and object commands, but uses UnifiedInput/UnifiedOutput. The storage session layer uses the unified context (unifiedBasicContext), and the callbacks go in libs/server/Storage/Functions/UnifiedStore/.

Blocking command RESP handler pattern

Blocking commands (BLPOP, BRPOP, BLMOVE, BLMPOP, BZPOPMIN, BZPOPMAX, BZMPOP) follow a distinct pattern. They do not use ref storageApi and instead interact with the CollectionItemBroker (libs/server/Objects/ItemBroker/CollectionItemBroker.cs), which manages blocking/waiting behavior:

private unsafe bool SortedSetBlockingPop(RespCommand command)
{
    if (parseState.Count < 2)
        return AbortWithWrongNumberOfArguments(command.ToString());

    if (!parseState.TryGetTimeout(parseState.Count - 1, out var timeout, out var error))
        return AbortWithErrorMessage(error);

    var keysBytes = new byte[parseState.Count - 1][];
    for (var i = 0; i < keysBytes.Length; i++)
        keysBytes[i] = parseState.GetArgSliceByRef(i).ToArray();

    var result = storeWrapper.itemBroker.GetCollectionItemAsync(command, keysBytes, this, timeout).Result;

    if (!result.Found)
    {
        WriteNull();
    }
    else
    {
        // Write RESP response with result.Key, result.Item, result.Score, etc.
    }

    return true;
}

Key differences from regular commands:

The dispatch in RespServerSession.cs does NOT pass ref storageApi: RespCommand.BZMPOP => SortedSetBlockingMPop(),
The handler calls storeWrapper.itemBroker.GetCollectionItemAsync() which blocks (with timeout) until data is available
No IGarnetApi method, no storage session method, and no RMW callbacks are needed for the blocking command itself (Steps 5-7 are skipped)
The CollectionItemBroker is notified when data is added to a collection (e.g., ZADD calls itemBroker.HandleCollectionUpdate(key)), which wakes up blocked clients
When adding a new blocking command, you must also update the TryGetResult method in CollectionItemBroker.cs to map your RespCommand to the correct GarnetObjectType and implement the retrieval logic

Steps 5–7: Storage Layer (skip for admin/non-key commands)

> Note: Steps 5, 6, and 7 apply only to commands that read or write key-value data through the store (e.g., SET, GET, DELIFGREATER). Admin commands like DEBUG, PING, CONFIG, etc. handle their logic entirely in the RESP handler (Step 4) and do not need API interface methods, storage session ops, or RMW callbacks. Skip to Step 8 for those. Blocking commands (e.g., BZMPOP) also skip Steps 5-7 — see the blocking command pattern in Step 4.

> Note on context types: The unified single-store has three context types: string context (for raw string commands like GET/SET), object context (for collection commands like ZADD/LPUSH), and unified context (for type-agnostic commands like EXISTS/DELETE/TTL/EXPIRE). Most new commands use either the string or object context — the unified context is only for commands that must work across both value types.

Step 5: Add API Interface Method

File: libs/server/API/IGarnetApi.cs

Add method signature to IGarnetApi (read-write) or IGarnetReadApi (read-only):

// String command:
GarnetStatus MyOperation(PinnedSpanByte key, ref StringInput input, ref StringOutput output);

// Object command:
GarnetStatus MyOperation(PinnedSpanByte key, ref ObjectInput input, ref GarnetObjectStoreOutput output);

// Unified command:
GarnetStatus MyOperation(PinnedSpanByte key, ref UnifiedInput input, ref UnifiedOutput output);

File: libs/server/API/GarnetApi*.cs

Add delegation in the GarnetApi partial struct. The implementation goes in the appropriate partial file based on the context type:

GarnetApi.cs — string commands
GarnetApiObjectCommands.cs — object commands
GarnetApiUnifiedCommands.cs — unified commands

public GarnetStatus MyOperation(PinnedSpanByte key, ref StringInput input, ref StringOutput output)
    => storageSession.MyOperation(key, ref input, ref output);

⚠️ Caveat: GarnetApi is a generic partial struct: GarnetApi<TStringContext, TObjectContext, TUnifiedContext>. Always add your method to the correct partial file for the context type you're using.

Overloads for programmatic callers: In addition to the primary signature (used by the network handler), you can add simpler overloads for programmatic/embedded callers that avoid forcing them to create the Input/Output structs. For example:

public GarnetStatus MyOperation(PinnedSpanByte key, double val, out double output)

This overload internally creates the appropriate input/output structs and only returns the desired value to the caller, instead of writing to the output buffer.

Step 6: Implement Storage Session Layer

File: New or existing file in libs/server/Storage/Session/MainStore/ (for string-context ops), libs/server/Storage/Session/ObjectStore/ (for object-context ops), or libs/server/Storage/Session/UnifiedStore/ (for unified-context ops)

String command pattern

This layer wraps Tsavorite API calls. The network path uses a generic context parameter:

public GarnetStatus MyOperation<TStringContext>(PinnedSpanByte key, ref StringInput input, ref StringOutput output, ref TStringContext context)
    where TStringContext : ITsavoriteContext<...>
{
    var status = context.RMW((FixedSpanByteKey)key, ref input, ref output);
    if (status.IsPending)
        CompletePendingForSession(ref status, ref output, ref context);

    return GarnetStatus.OK;
}

Object and unified commands follow the same pattern — just substitute the appropriate context, input, and output types:

|-------------|-----------|------------|---------------|

Programmatic overloads: You can also add simpler overloads for programmatic callers (see Step 5 note). These internally create the input/output structs and return only the desired value.

Object-specific: HandleCollectionUpdate

Object commands use the same pattern as above with ObjectInput/GarnetObjectStoreOutput and the object context.

⚠️ Caveat — HandleCollectionUpdate: If your object command modifies a collection (adds/removes elements), call itemBroker.HandleCollectionUpdate(key) after the store operation. This wakes up any clients blocked on that key (e.g., via BZPOPMIN). The actual data logic is implemented in the object class, not in the storage session.

Object implementation

File: libs/server/Objects/[ObjectName]/[ObjectName]ObjectImpl.cs

For object commands, the core logic lives in the object implementation. The Operate method in [ObjectName]Object.cs dispatches to implementation methods based on the operation enum:

case SortedSetOperation.ZADD:
    SortedSetAdd(ref input, ref output.SpanByteAndMemory);
    break;

The implementation methods in [ObjectName]ObjectImpl.cs directly manipulate the object's internal data structures (e.g., sortedSet, sortedSetDict for SortedSet).

Step 7: Add RMW/Read Callbacks (if applicable)

File: libs/server/Storage/Functions/MainStore/RMWMethods.cs (string commands) or libs/server/Storage/Functions/UnifiedStore/RMWMethods.cs (unified commands)

If your command uses RMW, you must handle these callbacks:

| Callback | When | Purpose |

|----------|------|---------|

| NeedInitialUpdate | Key doesn't exist | Return true to create record |

| InitialUpdater | Creating new record | Write initial value |

| NeedCopyUpdate | Key exists, record needs copy | Return true to copy-update, false to skip |

| InPlaceUpdater | Key exists, update in place | Modify existing value |

| CopyUpdater | Key exists, copy to new record | Write updated value to new record |

Add a case RespCommand.MYCMD: to each relevant switch statement.

For Read commands (MainStore): If your command uses Read (not RMW), the read response logic lives in libs/server/Storage/Functions/MainStore/PrivateMethods.cs — add a case to the CopyRespToWithInput method.

File: libs/server/Storage/Functions/MainStore/VarLenInputMethods.cs (string commands) or libs/server/Storage/Functions/UnifiedStore/VarLenInputMethods.cs (unified commands)

If your command writes a value, you must specify the value length:

| Method | Purpose |

|--------|---------|

| GetRMWInitialFieldInfo | Size of value for new records |

| GetRMWModifiedFieldInfo | Size of value for updated records |

⚠️ Caveat — RecordType:

If your command creates records with a custom RecordType (e.g., for type discrimination), you can set it in InitialUpdater after record initialization:

var header = logRecord.RecordDataHeader;
header.RecordType = MyManager.MyRecordType;

This works because RecordDataHeader.RecordType has a setter that writes through a raw pointer. No Tsavorite infrastructure changes needed (despite the TODO comments in LogRecord.cs).

Step 8: Command Metadata Registration

8a. Add to SupportedCommand.cs

File: playground/CommandInfoUpdater/SupportedCommand.cs

Add entry following the existing ordering/grouping in the file:

new("MY.CMD", RespCommand.MYCMD, StoreType.Main),

> Note: The file is not strictly alphabetical — entries are grouped by category (e.g., script commands at the end). Follow the existing grouping conventions rather than inserting strictly alphabetically.

For admin/non-key commands (e.g., DEBUG, PING), omit StoreType or use StoreType.None:

new("DEBUG", RespCommand.DEBUG),

StoreType values: Main (string store), Object (object store), All (both), None (no keys).

8b. Add to GarnetCommandsInfo.json (Garnet-only commands)

File: playground/CommandInfoUpdater/GarnetCommandsInfo.json

Needed for commands that don't exist in standard Redis (e.g., DELIFGREATER, SETIFMATCH), or standard Redis commands whose info you need to override. Standard Redis commands (e.g., DEBUG, GETDEL) normally get their metadata from a running RESP server automatically via the CommandInfoUpdater tool — skip this step and Step 8c for those unless you need to override their info.

> Overriding a command that also exists on the baseline server (e.g., OBJECT): the tool merges per sub-command, by name — your GarnetCommandsInfo.json / GarnetCommandsDocs.json entry wins and the baseline server fills in any sub-commands you did not specify. The override replaces the command's base fields wholesale (it is *not* a field-by-field merge), so include the parent command plus the sub-commands you want to control. This is how OBJECT reports Garnet-accurate sub-command summaries while the tool still generates the rest.

Add a JSON entry:

{
  "Command": "MYCMD",
  "Name": "MY.CMD",
  "IsInternal": false,
  "Arity": -2,
  "Flags": "DenyOom, Write",
  "FirstKey": 1,
  "LastKey": 1,
  "Step": 1,
  "AclCategories": "Slow, Write, Garnet",
  "KeySpecifications": [
    {
      "BeginSearch": { "TypeDiscriminator": "BeginSearchIndex", "Index": 1 },
      "FindKeys": { "TypeDiscriminator": "FindKeysRange", "LastKey": 0, "KeyStep": 1, "Limit": 0 },
      "Flags": "RW, Insert"
    }
  ],
  "StoreType": "Main"
}

Key fields:

Arity: Positive = exact arg count (including command name); Negative = minimum
Flags: ReadOnly, Write, DenyOom, Fast, Admin, NoAuth, Module, etc.
AclCategories: Used for ACL permission checks. Use Garnet for Garnet-specific commands
KeySpecifications: Drives automatic transaction key locking — no per-command switch needed
KeySpec.Flags: RO (read-only), RW (read-write), Access, Insert, Update, Delete

8c. Add to GarnetCommandsDocs.json (Garnet-only commands)

File: playground/CommandInfoUpdater/GarnetCommandsDocs.json

> Note: This step is not necessary for internal commands. The main purpose of command docs is to enable client auto-complete for the command.

Add documentation entry:

{
  "Command": "MYCMD",
  "Name": "MY.CMD",
  "Summary": "Description of what the command does.",
  "Group": "Generic",
  "Complexity": "O(1)",
  "Arguments": [
    {
      "TypeDiscriminator": "RespCommandKeyArgument",
      "Name": "KEY",
      "DisplayText": "key",
      "Type": "Key",
      "KeySpecIndex": 0
    }
  ]
}

⚠️ Caveat — Group must be a valid RespCommandGroup enum value:

Bitmap, Cluster, Connection, Generic, Geo, Hash, HyperLogLog, List, Module, PubSub, Scripting, Sentinel, Server, Set, SortedSet, Stream, String, Transactions

Do NOT invent new group names — the JSON deserializer will fail.

8d. Generate the resource JSON files

⚠️ CRITICAL: Never edit libs/resources/RespCommandsInfo.json or libs/resources/RespCommandsDocs.json directly. These are generated by the CommandInfoUpdater tool.

Steps:

Start a local RESP-compatible server (e.g., Valkey or Redis) — the tool queries it for standard Redis command metadata:

   valkey-server --port 6399

Build and run the tool:

   cd playground/CommandInfoUpdater
   dotnet build -f net10.0
   dotnet run -f net10.0 --no-build -- --port 6399 --output ../../libs/resources

(The --port must match the port of the local RESP server.)

The tool will prompt Would you like to continue? (Y/N) twice (once for info, once for docs). Press Y for both, or pass --yes (see below) to auto-confirm.
Kill the local RESP server afterward.

⚠️ Caveat: By default the tool prompts via Console.ReadKey(), which does NOT work with piped input (echo "Y" | dotnet run ... fails). For non-interactive / scripted runs (including AI agents), pass --yes to auto-confirm both prompts; otherwise run it in a real interactive terminal.

⚠️ Caveat: The tool requires a running RESP-compatible server (e.g., Valkey or Redis — not Garnet) to query standard command metadata. For Garnet-only commands, the tool reads from GarnetCommandsInfo.json and GarnetCommandsDocs.json instead.

⚠️ Caveat — unexpected "commands to remove": If the tool reports commands *to remove* that you did not intend to delete, those commands exist in the resource files but are missing from SupportedCommand.cs. Register them in SupportedCommand.cs (and, for Garnet-only commands, in the override JSONs) rather than reaching for the --ignore flag. Every command Garnet ships should be registered, so a normal run needs no --ignore.

> See also: playground/CommandInfoUpdater/README.md documents the tool in full — the inputs you edit, the per-sub-command override/merge behavior, the baseline server (Docker), the --ignore escape hatch, and how to surgically regenerate a single command.

Step 9: Add ACL Test

File: test/Garnet.test/Resp/ACL/RespCommandTests.cs

⚠️ CRITICAL: The AllCommandsCovered test automatically validates that every RespCommand enum value has a corresponding ACL test. If you add a command without an ACL test, AllCommandsCovered will fail.

Test naming convention:

Method name must end with ACLs or ACLsAsync
The name (minus the suffix) must match the command name with dots and underscores removed
Example: RI.CREATE → RICreateACLsAsync

Pattern (follow SADD for non-idempotent commands):

[Test]
public async Task MyCommandACLsAsync()
{
    int count = 0;

    await CheckCommandsAsync(
        "MY.CMD",
        [DoMyCommandAsync]
    ).ConfigureAwait(false);

    async Task DoMyCommandAsync(GarnetClient client)
    {
        var val = await client.ExecuteForStringResultAsync("MY.CMD",
            [$"key-{count}", "arg1"]).ConfigureAwait(false);
        count++;
        ClassicAssert.AreEqual("OK", val);
    }
}

⚠️ Caveat — Idempotency: The ACL framework calls your command multiple times (for different user/permission combinations). If your command is NOT idempotent (e.g., RI.CREATE fails on duplicate), use a counter to generate unique keys per invocation (see the SADD ACL test pattern).

Step 10: Add Integration Tests

File: Add tests to an existing or new file in test/Garnet.test/:

String / Unified commands: Add to test/Garnet.test/RespTests.cs
Object commands: Add to test/Garnet.test/Resp[ObjectName]Tests.cs (e.g., RespSortedSetTests.cs)
New feature area: Create test/Garnet.test/Resp<Feature>Tests.cs if the command doesn't fit existing test files

Required structure:

[TestFixture]
public class RespMyFeatureTests : TestBase
{
    GarnetServer server;

    [SetUp]
    public void Setup()
    {
        TestUtils.DeleteDirectory(TestUtils.MethodTestDir, wait: true);
        server = TestUtils.CreateGarnetServer(TestUtils.MethodTestDir);
        server.Start();
    }

    [TearDown]
    public void TearDown()
    {
        server.Dispose();
        TestUtils.OnTearDown();
    }

    [Test]
    public void MyBasicTest()
    {
        using var redis = ConnectionMultiplexer.Connect(TestUtils.GetConfig());
        var db = redis.GetDatabase(0);
        var result = db.Execute("MY.CMD", "key", "value");
        ClassicAssert.AreEqual("OK", (string)result);
    }
}

Note: Test fixtures must inherit from TestBase.

Recommended test cases:

Basic success case
Error cases (wrong args, wrong type)
Duplicate/idempotency behavior
DEL/UNLINK interaction (if applicable)
Type safety (e.g., string command on object key → WRONGTYPE)

Step 11: Update Website Documentation

File: website/docs/commands/ — choose the appropriate markdown file based on the command category (e.g., garnet-specific.md for Garnet-only commands, api-compatibility.md to mark a standard Redis command as supported).

Add a section documenting the command syntax, description, and response format:

### **MY.CMD**

#### **Syntax**

MY.CMD key value


Description of what the command does.

#### **Response**

- **Type reply**: Description of the response.

Also mark the command as supported in website/docs/commands/api-compatibility.md if it corresponds to a standard Redis command.

Step 11b: Add Configuration Settings (if needed)

If the command is optional, gated behind a feature flag, or needs a server-side configuration parameter (e.g., DEBUG requires --enable-debug-command), you must wire up a configuration option across four files:

1. Add property to `Options` class

File: libs/host/Configuration/Options.cs

Add a property with the [Option] attribute (from CommandLineParser):

[OptionValidation]
[Option("enable-my-feature", Required = false, HelpText = "Enable MY.CMD for 'no', 'local' or 'all' connections")]
public ConnectionProtectionOption EnableMyFeature { get; set; }

The [Option] attribute defines the CLI flag name (kebab-case). Use Required = false for optional settings. Common types: bool, int, string, ConnectionProtectionOption (for no/local/yes connection gating), or custom enums.

2. Map to `GarnetServerOptions`

File: libs/server/Servers/GarnetServerOptions.cs

Add a matching field:

/// <summary>
/// Enables MY.CMD
/// </summary>
public ConnectionProtectionOption EnableMyFeature;

Then in Options.GetServerOptions() (in Options.cs), map the property:

EnableMyFeature = EnableMyFeature,

3. Add default value

File: libs/host/defaults.conf

Add the default in the appropriate section:

/* Enable MY.CMD for clients - no/local/yes */
"EnableMyFeature": "no",

4. Check the setting in your RESP handler

Access the setting via storeWrapper.serverOptions:

if (storeWrapper.serverOptions.EnableMyFeature == ConnectionProtectionOption.No)
{
    while (!RespWriteUtils.TryWriteError("ERR command not enabled"u8, ref dcurr, dend))
        SendAndReset();
    return true;
}

5. Add config tests

File: test/Garnet.test/GarnetServerConfigTests.cs

Test that the setting is parsed correctly from CLI args and config files.

Step 12: Verify Everything

Build

dotnet build Garnet.slnx -c Debug

Format check

dotnet format Garnet.slnx --verify-no-changes

⚠️ Caveat: New files commonly fail with FINALNEWLINE errors. Ensure files do NOT have a trailing newline at the very end. Fix with: perl -pi -e 'chomp if eof' path/to/file.cs

Run your tests

dotnet test test/Garnet.test -f net10.0 -c Debug --filter "FullyQualifiedName~RespMyFeatureTests"

Run ACL coverage test

dotnet test test/Garnet.test -f net10.0 -c Debug --filter "FullyQualifiedName~AllCommandsCovered"

Run broader regression tests

dotnet test test/Garnet.test -f net10.0 -c Debug --filter "FullyQualifiedName~RespTests"

Transaction Support

For standard commands, transaction key locking is automatic — driven by KeySpecifications in the command metadata JSON. No per-command code is needed in TxnKeyManager.cs.

For custom multi-key operations that don't fit the standard key spec pattern, manually call txnManager.SaveKeyEntryToLock(key, lockType).

Common Caveats and Gotchas

Dot-prefixed commands (e.g., RI.CREATE): The RESP wire name uses a dot, but the enum name cannot. The ACL parser, AllCommandsCovered test, and SupportedCommand.cs all need to handle the dot-to-enum mapping. Check ACLParser.cs for normalization logic.

AllCommandsCovered is strict: It reflects over ALL RespCommand enum values and ALL entries in RespCommandsInfo.json. Missing either an ACL test or a JSON entry will fail this test.

NeedCopyUpdate for create-only commands: If your command should NOT overwrite existing records (like SETNX), return false from NeedCopyUpdate for your command. Otherwise Tsavorite will attempt a copy-update when the record can't be updated in-place.

VarLenInputMethods.cs is easy to forget: If your command creates or modifies records via RMW, you must add cases to GetRMWInitialFieldInfo and GetRMWModifiedFieldInfo. Without this, Tsavorite won't allocate the right amount of space for your value.

Resource JSON files are generated, not hand-edited: libs/resources/RespCommandsInfo.json and libs/resources/RespCommandsDocs.json are generated by playground/CommandInfoUpdater. Edit the source files (GarnetCommandsInfo.json, GarnetCommandsDocs.json, SupportedCommand.cs) and run the tool.

RespCommandGroup enum is closed: The Group field in docs JSON must match a value in the RespCommandGroup enum (libs/server/Resp/RespCommandDocs.cs). Use Generic if no existing group fits.

File headers: All C# files require // Copyright (c) Microsoft Corporation. / // Licensed under the MIT license.

Test resource usage: Use small values for buffer sizes, cache sizes, etc. in tests. Don't allocate 16MB+ buffers when 64KB will do.

Reference: RI.CREATE Implementation

The RI.CREATE command was the first command implemented following this guide. Key files for reference:

| File | Purpose |

|------|---------|

| libs/server/Resp/RangeIndex/RespServerSessionRangeIndex.cs | RESP handler with option parsing |

| libs/server/Resp/RangeIndex/RangeIndexManager.cs | Manager pattern for external data |

| libs/server/Resp/RangeIndex/RangeIndexManager.Index.cs | Fixed-size stub struct in store |

| libs/server/Storage/Session/MainStore/RangeIndexOps.cs | StorageSession → RMW flow |

| test/Garnet.test/RespRangeIndexTests.cs | Integration tests |

| test/Garnet.test/Resp/ACL/RespCommandTests.cs | ACL test (search for RICreateACLsAsync) |

How to use it

Copy the folder

Take microsoft/add-garnet-command from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.