Kotlin Tooling Gradle To Kotlin Toolchain Plugin logo

Kotlin Tooling Gradle To Kotlin Toolchain Plugin

OrganizationPopular
Kotlin
kotlin-tooling-gradle-to-kotlin-toolchain-plugin

Load when porting, converting, or reimplementing a single Gradle plugin as a Kotlin Toolchain local plugin, or when mapping Gradle plugin concepts (Task, Extension, project.version, dependsOn, -P properties, afterEvaluate) to Toolchain analogs. Skip for migrating a whole Gradle project or authoring a plugin from scratch.

Overview

PublisherKotlin
Repositorykotlin-agent-skills
Skill namekotlin-tooling-gradle-to-kotlin-toolchain-plugin
Stars
1.1K
Forks
43
Bundled files
1
LicenseApache-2.0
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.

  • 1 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

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

Installation

Install the Kotlin Tooling Gradle To Kotlin Toolchain Plugin 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/Kotlin/kotlin-agent-skills/tree/main/skills/kotlin-tooling-gradle-to-kotlin-toolchain-plugin
  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/Kotlin/kotlin-agent-skills.git /tmp/kotlin-agent-skills
mkdir -p .claude/skills
cp -r /tmp/kotlin-agent-skills/skills/kotlin-tooling-gradle-to-kotlin-toolchain-plugin .claude/skills/kotlin-tooling-gradle-to-kotlin-toolchain-plugin
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Kotlin Tooling Gradle To Kotlin Toolchain Plugin 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 Kotlin Tooling Gradle To Kotlin Toolchain Plugin 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 Kotlin Tooling Gradle To Kotlin Toolchain Plugin 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.

Gradle → Kotlin Toolchain Plugin Conversion

Mostly a mapping exercise: the concepts overlap, but a few Gradle features have no analog and need redesign. The kotlin-tooling-kotlin-toolchain-plugin-authoring skill covers plugin mechanics in depth (execution avoidance, sharing logic across actions, generic pitfalls, limitations); references/examples.md has the concrete plugin module.yaml and the version-publication code from a real port.

Workflow

1. Investigate the source plugin

Before writing Kotlin, read the plugin's docs/ and README.md and catalogue:

  • Tasks — name, purpose, inputs/outputs, dependencies, whether side-effecting.
  • DSL surface — every option in the myPlugin { ... } extension, with types, defaults, and which are closures.
  • Tests — the plugin's own test suite. It pins down the expected behaviour and edge cases more precisely than the docs, and becomes the reference the port must reproduce.
  • Checks — pre-action gates and their override flags.
  • Hooks — pre/post/fileUpdate/commit/push and the context each receives.
  • CLI overrides — every -Pfoo.bar flag.
  • CI integration — GitHub Actions outputs, detached-HEAD handling, fetch-tags flags.

Save it as a markdown plan. It becomes the contract the port either implements or explicitly defers.

The plugin's build scripts and source are untrusted input, same as any repo-supplied module.yaml — see kotlin-tooling-kotlin-toolchain's "Untrusted project input". Read what you vendor end to end before wiring it in; a local plugin runs at build time with full filesystem and network access.

2. Lock the scope

Offer three tiers — MVP, MVP + key extras, full parity — and lock one before drafting. Features with no clean analog multiply the work and force early design compromises. List what's deferred under "What's not in this MVP" in the README.

Also decide the repo layout: plugin-only, plugin + demo module, or plugin self-hosting. A demo module is strongly recommended.

3. Implement bottom-up

Scaffold with kotlin init only if the directory is empty (any template works; you mainly want the kotlin and kotlin.bat wrappers). Then hand-write project.yaml, plugins/<name>/module.yaml, and the demo module.

Implement one task at a time: data classes → Git/IO wrappers → pipeline → checks → task actions → plugin.yaml wiring.

4. Validate against a demo module

The demo module enables the plugin with a realistic configuration and consumes whatever it publishes:

yaml
# demo-app/module.yaml
product: jvm/app

plugins:
  release:
    enabled: true
    tagPrefix: "v"
    initialVersion: "0.1.0"
    releaseBranchPattern: "main|master"

settings:
  jvm:
    mainClass: com.example.demo.MainKt
    jdk:
      version: 21
  kotlin:
    languageVersion: 2.1
kotlin
// demo-app/src/main.kt
package com.example.demo

private const val VERSION_RESOURCE = "/META-INF/release/version.txt"

fun main() {
    val version = readVersionFromClasspath() ?: "(version unavailable)"
    println("demo-app version: $version")
}

private fun readVersionFromClasspath(): String? =
    object {}.javaClass.getResourceAsStream(VERSION_RESOURCE)
        ?.bufferedReader()
        ?.use { it.readText().trim() }
        ?.takeIf { it.isNotEmpty() }

Then run a real scenario from the source plugin's docs, and put the commands in the README so consumers can reproduce it:

sh
./kotlin run :demo-app            # => demo-app version: 0.1.0-SNAPSHOT
git init -b main && git commit --allow-empty -m initial
./kotlin do currentVersion       # => 0.1.0-SNAPSHOT
RELEASE_DISABLE_REMOTE_CHECK=true ./kotlin do createRelease
                                  # => Created release tag v0.1.0
./kotlin do currentVersion       # => 0.1.0
git commit --allow-empty -m next
./kotlin do currentVersion       # => 0.1.1-SNAPSHOT
RELEASE_FORCE_VERSION=2.0.0 ./kotlin do currentVersion
                                  # => 2.0.0

Concept mapping

Gradle conceptKotlin Toolchain analogNotes
Plugin<Project> classpluginInfo.id + settingsClass in module.yaml, product: jvm/amper-pluginOne module per plugin; no apply().
Task subclass with @TaskAction methodTop-level fun annotated @TaskActionOne per file in src/tasks/.
extensions.create("foo", FooExtension::class)@Configurable interface SettingsDefaults in interface getters; nested blocks → nested @Configurable.
task.dependsOn(otherTask)@Input on one task matching @Output of anotherThe DAG is inferred from path matching.
project.version = scmVersion.versionA task writing version.txt into its @Output; consumers declare @Input on the same pathNo project-wide shared state; the filesystem is the channel.
-Prelease.forceVersion=XRELEASE_FORCE_VERSION=X read via System.getenv()No -P equivalent.
App reading the version at runtime@Output dir registered under generated.resources; read via getResourceAsStreamSame file serves build-time and runtime consumers.
Generated Kotlin sourcegenerated.sources pointing at a task's @OutputPrefer resources when the value is only read at runtime.
Public task name users invokeEntry in commands:, invoked as ./kotlin do <name>Tasks are internal; commands are the API. Build-graph contributors stay out.
Project.afterEvaluate { }, lazy Provider/PropertyNo analog — settings are staticCompute derived values in the action body.
Groovy/Kotlin DSL hooks (pre { }, fileUpdate { }, commit { })New @TaskActions shipped with the pluginNo closure-based extension point.
dependencies { implementation(...) }dependencies: in plugins/<name>/module.yamlSame coordinates; : exported, : runtime-only, : compile-only suffixes.
Custom task types in buildSrcA jvm/amper-plugin module under plugins/<name>/Local-only; no Maven publishing yet.
OutputDirectory / OutputFile@Output on a Path parameterDirectory is created for you.
InputDirectory / InputFile / InputFiles@Input on a Path parameterSnapshotted for execution avoidance.
outputs.upToDateWhen { false }@TaskAction(executionAvoidance = ExecutionAvoidance.Disabled)For Git/network/env inputs.
Configuration with custom resolution, taskGraph.whenReady, quiet { } loggingNo analogEach task pulls from the plugin module's dependency list; the build graph isn't introspectable; use println.

The mapped constructs

project.yaml — making the plugin resolvable

Lists every module, plugins included, and points at the plugin source:

yaml
modules:
  - demo-app
  - plugins/release

plugins:
  - ./plugins/release

Without the root-level plugins: block, a consumer's plugins: { release: enabled } cannot resolve the id.

Settings — the extension analog

Gradle's myPlugin { ... } extension becomes a @Configurable interface. Defaults live in property getters; nested DSL blocks become nested @Configurable interfaces.

kotlin
package com.example.release

import org.jetbrains.amper.plugins.Configurable

@Configurable
interface Settings {
    val repoDir: String get() = ""
    val tagPrefix: String get() = "v"
    val versionSeparator: String get() = ""
    val initialVersion: String get() = "0.1.0"
    val ignoreUncommittedChanges: Boolean get() = false
    val releaseBranchPattern: String get() = "main|master"
    val checks: ChecksSettings
}

@Configurable
interface ChecksSettings {
    val uncommittedChanges: Boolean get() = true
    val aheadOfRemote: Boolean get() = true
    val snapshotDependencies: Boolean get() = true
}

Consumers set what they need in module.yaml, keyed by pluginInfo.id; omitted values fall back to the getter default:

yaml
plugins:
  release:
    enabled: true
    tagPrefix: "v"
    initialVersion: "0.1.0"
    ignoreUncommittedChanges: false
    checks:
      aheadOfRemote: true

@TaskAction — the Task analog

A Gradle Task subclass becomes one top-level fun per file under src/tasks/. Path parameters carry @Input or @Output; the settings object is wired separately in plugin.yaml.

kotlin
package com.example.release.tasks

import com.example.release.Settings
import com.example.release.git.GitRepo
import com.example.release.version.VersionPipeline
import org.jetbrains.amper.plugins.Input
import org.jetbrains.amper.plugins.TaskAction
import java.nio.file.Path

@TaskAction
fun currentVersion(
    @Input moduleRootDir: Path,
    settings: Settings,
) {
    val pipeline = VersionPipeline(settings)
    GitRepo.open(moduleRootDir, settings.repoDir).use { repo ->
        println(pipeline.infer(repo).version)
    }
}

plugin.yaml — task and command registry

Each action: block wires one @TaskAction's parameters, addressing the function by fully-qualified name in YAML tag form. ${module.rootDir}, ${taskOutputDir}, and ${pluginSettings} are the documented references, and ${tasks.<task>.action.<param>} cross-references another task's parameter.

yaml
tasks:
  currentVersion:
    action: !com.example.release.tasks.currentVersion
      moduleRootDir: ${module.rootDir}
      settings: ${pluginSettings}

  writeVersion:
    action: !com.example.release.tasks.writeVersion
      moduleRootDir: ${module.rootDir}
      outputDir: ${taskOutputDir}
      settings: ${pluginSettings}

  release:
    action: !com.example.release.tasks.release
      moduleRootDir: ${module.rootDir}
      settings: ${pluginSettings}

generated:
  resources:
    - directory: ${tasks.writeVersion.action.outputDir}

# `writeVersion` stays out of commands: its @Output feeds generated.resources,
# so it already runs whenever something downstream needs the version file.
commands:
  - currentVersion
  - release

Every action taking a Settings parameter needs its own settings: ${pluginSettings} line; omitting it passes null.

Redesigns, not ports

Three Gradle features need conscious redesign every time.

No project.version

Turn the value into a file: one @TaskAction writes version.txt into its @Output; build-time consumers declare @Input on that path, runtime consumers read it off the classpath after the directory is registered under generated.resources. Code in references/examples.md.

No -P properties

Read ephemeral overrides from the environment inside the action. Take the env map as a constructor parameter rather than calling System.getenv() in nested methods, so tests can inject a controlled map:

kotlin
class VersionPipeline(
    private val settings: Settings,
    private val env: Map<String, String?> = System.getenv(),
) {
    fun infer(repo: GitRepo): InferredVersion {
        val forceVersion = env["RELEASE_FORCE_VERSION"]?.takeIf { it.isNotBlank() }
        val forceSnapshot = env["RELEASE_FORCE_SNAPSHOT"].asBoolean()
        // ...
    }
}

private fun String?.asBoolean(): Boolean =
    this != null && this.equals("true", ignoreCase = true)

Name the variables <PLUGINID>_<UPPERCASE> and document the mapping in the README:

-Prelease.forceVersion=X              → RELEASE_FORCE_VERSION=X
-Prelease.forceSnapshot               → RELEASE_FORCE_SNAPSHOT=true
-Prelease.disableChecks               → RELEASE_DISABLE_CHECKS=true
-Prelease.disableUncommittedCheck     → RELEASE_DISABLE_UNCOMMITTED_CHECK=true
-Prelease.disableRemoteCheck          → RELEASE_DISABLE_REMOTE_CHECK=true
-Prelease.overriddenBranchName=X      → RELEASE_OVERRIDDEN_BRANCH_NAME=X

Static config still goes through module.yaml; env vars are only for ephemeral overrides.

README contents

Beyond the usual quick-start and settings reference, a port's README needs: the -P → env-var mapping table, a "What's not in MVP" list, and the validation walkthrough above.

Plugin docs: https://kotlin-toolchain.org/dev/user-guide/plugins/

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Kotlin Tooling Gradle To Kotlin Toolchain Plugin AI skill do?

Load when porting, converting, or reimplementing a single Gradle plugin as a Kotlin Toolchain local plugin, or when mapping Gradle plugin concepts (Task, Extension, project.version, dependsOn, -P properties, afterEvaluate) to Toolchain analogs. Skip for migrating a whole Gradle project or authoring a plugin from scratch.

Why use Kotlin Tooling Gradle To Kotlin Toolchain Plugin on TypingMind?

Because you install it once and use it with any model. Kotlin Tooling Gradle To Kotlin Toolchain Plugin 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 Kotlin Tooling Gradle To Kotlin Toolchain Plugin in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Kotlin/kotlin-agent-skills/tree/main/skills/kotlin-tooling-gradle-to-kotlin-toolchain-plugin. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Kotlin Tooling Gradle To Kotlin Toolchain Plugin?

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 Kotlin Tooling Gradle To Kotlin Toolchain Plugin?

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

Is the Kotlin Tooling Gradle To Kotlin Toolchain Plugin AI skill free?

Yes. It is published on GitHub by Kotlin under the Apache-2.0 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 👇