Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

Migrating to Modifier.Node — Persistent Nodes Over composed { } · skydoves | DecimalAI

Frontendby skydoves

Migrating to Modifier.Node — Persistent Nodes Over composed { }

Use this skill to author new custom Jetpack Compose modifiers and migrate legacy ones from Modifier.composed { } to Modifier.Node + ModifierNodeElement<T>. Covers the persistent-node lifecycle (onAttach, onDetach, onReset, coroutineScope), the specialized node interfaces (DrawModifierNode, LayoutModifierNode, SemanticsModifierNode, PointerInputModifierNode, CompositionLocalConsumerModifierNode, LayoutAwareModifierNode, GlobalPositionAwareModifierNode, ObserverModifierNode, DelegatingNode, Traver

Never made a case worseMaintained

SkillScore

SkillScore88/100

Test results 91 · Live runs 86

Safety

3/3 safety checks passed

Code scan
Intent
Content

What it changes

82%91%of test cases passed +9 pts

Without skill18 of 22

With skill20 of 22

2 fixed18 passed either way2 still failing0 broke

Measured on gemini-3.6-flash8/19/2026v1

Track record

Agent runs: 15; 15 in the last 30 days

See one task, with and without the skill

One test taskIn a performance-critical custom node, we want to update internal parameter state in `update(node)` without triggering the framework's automatic full invalidation pass, opting to trigger drawing invalidation manually. Which property and method should be used?

Without the skillMissed

With the skillPassed

Same model, same task. Answers are capped and redacted.

Use this skill

Your agent runs on

Adds every file the skill has, pinned so you can update it. No account needed.

pip install "decimalai>=0.15"
decimalai skills add skydoves-migrating-to-modifier-node
claude --plugin-dir "$(git rev-parse --show-toplevel 2>/dev/null || pwd)/.agents"

The same install from Python, for a notebook or a setup script.

Install

pip install "decimalai>=0.15"

Add the skill

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("skydoves-migrating-to-modifier-node")
if result.exit_code:
    raise SystemExit(result.problems)

print("Added. Start Claude Code with the command below.")

Start Claude Code with it

claude --plugin-dir "$(git rev-parse --show-toplevel 2>/dev/null || pwd)/.agents"

Type these in a Claude Code session. No account or API key. It writes the skill and any files it bundles, and asks first when its safety rating isn't Passed.

Add the DecimalAI plugin, once

/plugin marketplace add decimal-labs/decimalai-claude-plugins
/plugin install decimalai@decimalai

Install this skill

/decimalai:install skydoves-migrating-to-modifier-node

The plugin needs python3 and curl on your machine.

No Python or account. Claude Code picks it up in its next session.

Save it into your project

curl -fsSL --create-dirs -o .claude/skills/skydoves-migrating-to-modifier-node/SKILL.md https://app.decimal.ai/s/skydoves-migrating-to-modifier-node/SKILL.md

Copy SKILL.md

Save it in your project, as

.claude/skills/skydoves-migrating-to-modifier-node/SKILL.md

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Claude Code docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[claude-agent-sdk]"
decimalai skills add skydoves-migrating-to-modifier-node

Add it to your agent

import asyncio

from claude_agent_sdk import ClaudeAgentOptions
from decimalai.claude_agent_sdk import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

async def main():
    question = "Hello!"
    async with support.run(question) as run:  # one user message, one run
        async for message in run.query(options=ClaudeAgentOptions(max_turns=6)):
            print(message)

# In a notebook or async code: await main()
asyncio.run(main())

Prefer its own loader? ClaudeAgentOptions(plugins=[skills_plugin()], skills="all") reads the same folder. Use one or the other: a skill delivered both ways reaches the prompt twice.

The same install from Python, for a notebook or a setup script. agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[claude-agent-sdk]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("skydoves-migrating-to-modifier-node")
if result.exit_code:
    raise SystemExit(result.problems)

import asyncio

from claude_agent_sdk import ClaudeAgentOptions
from decimalai.claude_agent_sdk import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

async def main():
    question = "Hello!"
    async with support.run(question) as run:  # one user message, one run
        async for message in run.query(options=ClaudeAgentOptions(max_turns=6)):
            print(message)

# In a notebook or async code: await main()
asyncio.run(main())

Prefer its own loader? ClaudeAgentOptions(plugins=[skills_plugin()], skills="all") reads the same folder. Use one or the other: a skill delivered both ways reaches the prompt twice.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/skydoves-migrating-to-modifier-node/SKILL.md

Give it to your agent

import asyncio
from pathlib import Path
from claude_agent_sdk import ClaudeAgentOptions, query

skill = Path(".agents/skills/skydoves-migrating-to-modifier-node/SKILL.md").read_text()
options = ClaudeAgentOptions(system_prompt=f"You are a helpful agent.\n\n{skill}", max_turns=6)

async def main():
    async for message in query(prompt="Hello!", options=options):
        print(message)

# In a notebook or async code: await main()
asyncio.run(main())

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Claude Agent SDK docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[openai-agents]"
decimalai skills add skydoves-migrating-to-modifier-node

Add it to your agent

from agents import Agent, Runner
from decimalai.openai_agents import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = Agent(
    name=support.name,
    instructions=support.instructions,  # your prompt, then the skill list
    tools=[*support.tools],             # your tools, then load_skill
)
# In a notebook or async code: result = await Runner.run(agent, "Hello!")
result = Runner.run_sync(agent, "Hello!")

Using SandboxAgent? capabilities=[Shell(), decimalai.openai_agents.sandbox_skills()] points its own loader at the same folder.

Install

pip install "decimalai[openai-agents]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("skydoves-migrating-to-modifier-node")
if result.exit_code:
    raise SystemExit(result.problems)

from agents import Agent, Runner
from decimalai.openai_agents import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = Agent(
    name=support.name,
    instructions=support.instructions,  # your prompt, then the skill list
    tools=[*support.tools],             # your tools, then load_skill
)
# In a notebook or async code: result = await Runner.run(agent, "Hello!")
result = Runner.run_sync(agent, "Hello!")

Using SandboxAgent? capabilities=[Shell(), decimalai.openai_agents.sandbox_skills()] points its own loader at the same folder.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/skydoves-migrating-to-modifier-node/SKILL.md

Give it to your agent

from pathlib import Path
from agents import Agent, Runner

skill = Path(".agents/skills/skydoves-migrating-to-modifier-node/SKILL.md").read_text()
agent = Agent(name="support", instructions=f"You are a helpful agent.\n\n{skill}")
# In a notebook or async code: result = await Runner.run(agent, "Hello!")
result = Runner.run_sync(agent, "Hello!")

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

OpenAI Agents docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[openai]"
decimalai skills add skydoves-migrating-to-modifier-node

Add it to your agent

import json

from openai import OpenAI
from decimalai.openai import agent_config

client = OpenAI()
# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")


def lookup_order(order_id: str) -> str:
    return f"Order {order_id} shipped yesterday."


# Your tools, as Chat Completions takes them, and what answers each.
TOOLS = [{"type": "function", "function": {
    "name": "lookup_order",
    "description": "Look up the status of an order.",
    "parameters": {"type": "object", "properties": {"order_id": {"type": "string"}},
                   "required": ["order_id"]},
}}]
FUNCTIONS = {"lookup_order": lookup_order}

question = "Where is order A1001?"
with support.run(question) as run:  # one user message, one run
    # run.system is your prompt, then the skill list.
    messages = [{"role": "system", "content": run.system}, run.message]
    while True:
        reply = client.chat.completions.create(
            model="gpt-5.4-mini",
            messages=messages,
            tools=[*TOOLS, *run.tools],  # your tools, then load_skill
        ).choices[0].message
        messages.append(reply)
        if not reply.tool_calls:
            break
        for call in reply.tool_calls:
            # load_skill is answered by the run; your own tools by FUNCTIONS.
            function = FUNCTIONS.get(call.function.name)
            messages.append(run.tool_result(call) or {
                "role": "tool", "tool_call_id": call.id,
                "content": function(**json.loads(call.function.arguments)) if function
                else f"There is no tool named {call.function.name!r}.",
            })
print(reply.content)

# The Responses API takes run.response_tools instead of run.tools.

Install

pip install "decimalai[openai]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("skydoves-migrating-to-modifier-node")
if result.exit_code:
    raise SystemExit(result.problems)

import json

from openai import OpenAI
from decimalai.openai import agent_config

client = OpenAI()
# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")


def lookup_order(order_id: str) -> str:
    return f"Order {order_id} shipped yesterday."


# Your tools, as Chat Completions takes them, and what answers each.
TOOLS = [{"type": "function", "function": {
    "name": "lookup_order",
    "description": "Look up the status of an order.",
    "parameters": {"type": "object", "properties": {"order_id": {"type": "string"}},
                   "required": ["order_id"]},
}}]
FUNCTIONS = {"lookup_order": lookup_order}

question = "Where is order A1001?"
with support.run(question) as run:  # one user message, one run
    # run.system is your prompt, then the skill list.
    messages = [{"role": "system", "content": run.system}, run.message]
    while True:
        reply = client.chat.completions.create(
            model="gpt-5.4-mini",
            messages=messages,
            tools=[*TOOLS, *run.tools],  # your tools, then load_skill
        ).choices[0].message
        messages.append(reply)
        if not reply.tool_calls:
            break
        for call in reply.tool_calls:
            # load_skill is answered by the run; your own tools by FUNCTIONS.
            function = FUNCTIONS.get(call.function.name)
            messages.append(run.tool_result(call) or {
                "role": "tool", "tool_call_id": call.id,
                "content": function(**json.loads(call.function.arguments)) if function
                else f"There is no tool named {call.function.name!r}.",
            })
print(reply.content)

# The Responses API takes run.response_tools instead of run.tools.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/skydoves-migrating-to-modifier-node/SKILL.md

Give it to your agent

from pathlib import Path
from openai import OpenAI

client = OpenAI()
skill = Path(".agents/skills/skydoves-migrating-to-modifier-node/SKILL.md").read_text()
reply = client.chat.completions.create(
    model="gpt-5.4-mini",
    messages=[
        {"role": "system", "content": f"You are a helpful agent.\n\n{skill}"},
        {"role": "user", "content": "Hello!"},
    ],
).choices[0].message

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

OpenAI SDK docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[anthropic]"
decimalai skills add skydoves-migrating-to-modifier-node

Add it to your agent

import anthropic
from decimalai.anthropic import agent_config

client = anthropic.Anthropic()
# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

question = "Hello!"
with support.run(question) as run:  # one user message, one run
    answer = client.beta.messages.tool_runner(
        model="claude-haiku-5-5",
        max_tokens=4096,
        system=run.system,     # your prompt, then the skill list
        tools=[*run.tools],    # your tools, then load_skill
        messages=[{"role": "user", "content": question}],
    ).until_done()

Install

pip install "decimalai[anthropic]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("skydoves-migrating-to-modifier-node")
if result.exit_code:
    raise SystemExit(result.problems)

import anthropic
from decimalai.anthropic import agent_config

client = anthropic.Anthropic()
# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

question = "Hello!"
with support.run(question) as run:  # one user message, one run
    answer = client.beta.messages.tool_runner(
        model="claude-haiku-5-5",
        max_tokens=4096,
        system=run.system,     # your prompt, then the skill list
        tools=[*run.tools],    # your tools, then load_skill
        messages=[{"role": "user", "content": question}],
    ).until_done()

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/skydoves-migrating-to-modifier-node/SKILL.md

Give it to your agent

from pathlib import Path
import anthropic

client = anthropic.Anthropic()
skill = Path(".agents/skills/skydoves-migrating-to-modifier-node/SKILL.md").read_text()
reply = client.messages.create(
    model="claude-haiku-5-5",
    max_tokens=4096,
    system=f"You are a helpful agent.\n\n{skill}",
    messages=[{"role": "user", "content": "Hello!"}],
)

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Anthropic docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[google-genai]"
decimalai skills add skydoves-migrating-to-modifier-node

Add it to your agent

from google import genai
from google.genai import types
from decimalai.google_genai import agent_config

client = genai.Client()
# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")


def lookup_order(order_id: str) -> str:
    return f"Order {order_id} shipped yesterday."


# Your functions, as declarations for the model, and what answers each.
TOOLS = [{"function_declarations": [{
    "name": "lookup_order",
    "description": "Look up the status of an order.",
    "parameters_json_schema": {"type": "object", "properties": {"order_id": {"type": "string"}},
                               "required": ["order_id"]},
}]}]
FUNCTIONS = {"lookup_order": lookup_order}

question = "Where is order A1001?"
with support.run(question) as run:  # one user message, one run
    config = types.GenerateContentConfig(
        system_instruction=run.system,     # your prompt, then the skill list
        tools=[*TOOLS, *run.tool_params],  # your functions, then load_skill
    )
    contents = [{"role": "user", "parts": [{"text": run.message}]}]
    while True:
        response = client.models.generate_content(
            model="gemini-3.5-flash", contents=contents, config=config,
        )
        if not response.function_calls:
            break
        contents.append(response.candidates[0].content)
        parts = []
        for call in response.function_calls:
            # load_skill is answered by the run; your own functions by FUNCTIONS.
            # A loop of your own rather than automatic function calling, which
            # raises KeyError when the model asks for a function you did not pass.
            function = FUNCTIONS.get(call.name)
            parts.append(run.tool_result(call) or {"function_response": {
                "name": call.name, "id": call.id,
                "response": {"result": function(**(call.args or {})) if function
                             else f"There is no function named {call.name!r}."},
            }})
        contents.append({"role": "user", "parts": parts})
print(response.text)

Install

pip install "decimalai[google-genai]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("skydoves-migrating-to-modifier-node")
if result.exit_code:
    raise SystemExit(result.problems)

from google import genai
from google.genai import types
from decimalai.google_genai import agent_config

client = genai.Client()
# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")


def lookup_order(order_id: str) -> str:
    return f"Order {order_id} shipped yesterday."


# Your functions, as declarations for the model, and what answers each.
TOOLS = [{"function_declarations": [{
    "name": "lookup_order",
    "description": "Look up the status of an order.",
    "parameters_json_schema": {"type": "object", "properties": {"order_id": {"type": "string"}},
                               "required": ["order_id"]},
}]}]
FUNCTIONS = {"lookup_order": lookup_order}

question = "Where is order A1001?"
with support.run(question) as run:  # one user message, one run
    config = types.GenerateContentConfig(
        system_instruction=run.system,     # your prompt, then the skill list
        tools=[*TOOLS, *run.tool_params],  # your functions, then load_skill
    )
    contents = [{"role": "user", "parts": [{"text": run.message}]}]
    while True:
        response = client.models.generate_content(
            model="gemini-3.5-flash", contents=contents, config=config,
        )
        if not response.function_calls:
            break
        contents.append(response.candidates[0].content)
        parts = []
        for call in response.function_calls:
            # load_skill is answered by the run; your own functions by FUNCTIONS.
            # A loop of your own rather than automatic function calling, which
            # raises KeyError when the model asks for a function you did not pass.
            function = FUNCTIONS.get(call.name)
            parts.append(run.tool_result(call) or {"function_response": {
                "name": call.name, "id": call.id,
                "response": {"result": function(**(call.args or {})) if function
                             else f"There is no function named {call.name!r}."},
            }})
        contents.append({"role": "user", "parts": parts})
print(response.text)

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/skydoves-migrating-to-modifier-node/SKILL.md

Give it to your agent

from pathlib import Path
from google import genai
from google.genai import types

client = genai.Client()
skill = Path(".agents/skills/skydoves-migrating-to-modifier-node/SKILL.md").read_text()
response = client.models.generate_content(
    model="gemini-3.5-flash",
    contents="Hello!",
    config=types.GenerateContentConfig(system_instruction=f"You are a helpful agent.\n\n{skill}"),
)

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Gemini API docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[langchain]" langchain-openai
decimalai skills add skydoves-migrating-to-modifier-node

Add it to your agent

from langchain.agents import create_agent
from decimalai.langchain import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = create_agent(
    "openai:gpt-5.4-mini",
    tools=[],                         # your tools; the middleware adds load_skill
    middleware=[support.middleware],  # your prompt, the skill list and load_skill
)
result = agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

Install

pip install "decimalai[langchain]" langchain-openai

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("skydoves-migrating-to-modifier-node")
if result.exit_code:
    raise SystemExit(result.problems)

from langchain.agents import create_agent
from decimalai.langchain import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = create_agent(
    "openai:gpt-5.4-mini",
    tools=[],                         # your tools; the middleware adds load_skill
    middleware=[support.middleware],  # your prompt, the skill list and load_skill
)
result = agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/skydoves-migrating-to-modifier-node/SKILL.md

Give it to your agent

from pathlib import Path
from langchain.agents import create_agent

skill = Path(".agents/skills/skydoves-migrating-to-modifier-node/SKILL.md").read_text()
agent = create_agent("openai:gpt-5.4-mini", tools=[], system_prompt=f"You are a helpful agent.\n\n{skill}")
result = agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

LangChain docs

CrewAI's own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" "crewai>=1.15.27"
decimalai skills add skydoves-migrating-to-modifier-node

Add it to your agent

import decimalai.crewai
from crewai import Agent, Crew, Task

agent = Agent(
    role="Support", goal="Help the customer", backstory="A support agent.",
    llm=llm,                               # your own CrewAI LLM
    skills=decimalai.crewai.skills(),      # the installed folder
    tools=decimalai.crewai.skill_tools(),  # reads a skill's files, runs its scripts
)
task = Task(description="Hello!", expected_output="an answer", agent=agent)
Crew(agents=[agent], tasks=[task]).kickoff()

The same install from Python, for a notebook or a setup script. CrewAI's own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" "crewai>=1.15.27"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("skydoves-migrating-to-modifier-node")
if result.exit_code:
    raise SystemExit(result.problems)

import decimalai.crewai
from crewai import Agent, Crew, Task

agent = Agent(
    role="Support", goal="Help the customer", backstory="A support agent.",
    llm=llm,                               # your own CrewAI LLM
    skills=decimalai.crewai.skills(),      # the installed folder
    tools=decimalai.crewai.skill_tools(),  # reads a skill's files, runs its scripts
)
task = Task(description="Hello!", expected_output="an answer", agent=agent)
Crew(agents=[agent], tasks=[task]).kickoff()

No DecimalAI package. CrewAI's own loader reads the folder, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/skydoves-migrating-to-modifier-node/SKILL.md

Give it to your agent

from pathlib import Path
from crewai import Agent, Crew, Task

agent = Agent(
    role="Support", goal="Help the customer", backstory="A support agent.",
    llm=llm,                         # your own CrewAI LLM
    skills=[Path(".agents/skills")],
)
task = Task(description="Hello!", expected_output="an answer", agent=agent)
Crew(agents=[agent], tasks=[task]).kickoff()

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

CrewAI docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[adk]"
decimalai skills add skydoves-migrating-to-modifier-node

Add it to your agent

from decimalai.adk import DecimalAIPlugin, agent_config
from google.adk.agents import LlmAgent
from google.adk.apps import App

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = LlmAgent(
    name="support",          # ADK needs a Python identifier
    model="gemini-3.5-flash",
    instruction=support.instruction,  # your prompt, then the skill list
    tools=[*support.tools],           # your tools, then load_skill
)
# DecimalAIPlugin makes each invocation of runners built from this App one trace.
app = App(name="support", root_agent=agent, plugins=[DecimalAIPlugin()])

Prefer ADK's own loader? SkillToolset(skills=decimalai.adk.skills(), ...) reads the same folder. Use one or the other: a skill delivered both ways reaches the prompt twice.

Install

pip install "decimalai[adk]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("skydoves-migrating-to-modifier-node")
if result.exit_code:
    raise SystemExit(result.problems)

from decimalai.adk import DecimalAIPlugin, agent_config
from google.adk.agents import LlmAgent
from google.adk.apps import App

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = LlmAgent(
    name="support",          # ADK needs a Python identifier
    model="gemini-3.5-flash",
    instruction=support.instruction,  # your prompt, then the skill list
    tools=[*support.tools],           # your tools, then load_skill
)
# DecimalAIPlugin makes each invocation of runners built from this App one trace.
app = App(name="support", root_agent=agent, plugins=[DecimalAIPlugin()])

Prefer ADK's own loader? SkillToolset(skills=decimalai.adk.skills(), ...) reads the same folder. Use one or the other: a skill delivered both ways reaches the prompt twice.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/skydoves-migrating-to-modifier-node/SKILL.md

Give it to your agent

from pathlib import Path
from google.adk.agents import LlmAgent

skill = Path(".agents/skills/skydoves-migrating-to-modifier-node/SKILL.md").read_text()
agent = LlmAgent(
    name="support",
    model="gemini-3.5-flash",
    instruction=lambda context: f"You are a helpful agent.\n\n{skill}",
)

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Google ADK docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[pydantic-ai]"
decimalai skills add skydoves-migrating-to-modifier-node

Add it to your agent

from pydantic_ai import Agent
from decimalai.pydantic_ai import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = Agent(
    "openai:gpt-4o-mini",
    name=support.name,
    tools=[],                 # your tools; the capability adds load_skill
    capabilities=[support],   # your prompt, the skill list and load_skill
)
# In a notebook or async code: result = await agent.run("Hello!")
result = agent.run_sync("Hello!")

On the Pydantic AI harness, decimalai.pydantic_ai_harness.skills(project) uses its own loader instead.

Install

pip install "decimalai[pydantic-ai]"

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("skydoves-migrating-to-modifier-node")
if result.exit_code:
    raise SystemExit(result.problems)

from pydantic_ai import Agent
from decimalai.pydantic_ai import agent_config

# With no DECIMAL_API_KEY this reads the skills your project lists, from disk.
support = agent_config("support", fallback="You are a helpful agent.")

agent = Agent(
    "openai:gpt-4o-mini",
    name=support.name,
    tools=[],                 # your tools; the capability adds load_skill
    capabilities=[support],   # your prompt, the skill list and load_skill
)
# In a notebook or async code: result = await agent.run("Hello!")
result = agent.run_sync("Hello!")

On the Pydantic AI harness, decimalai.pydantic_ai_harness.skills(project) uses its own loader instead.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/skydoves-migrating-to-modifier-node/SKILL.md

Give it to your agent

from pathlib import Path
from pydantic_ai import Agent

skill = Path(".agents/skills/skydoves-migrating-to-modifier-node/SKILL.md").read_text()
agent = Agent("openai:gpt-4o-mini", instructions=f"You are a helpful agent.\n\n{skill}")
# In a notebook or async code: result = await agent.run("Hello!")
result = agent.run_sync("Hello!")

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Pydantic AI docs

Deep Agents' own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" deepagents
decimalai skills add skydoves-migrating-to-modifier-node

Add it to your agent

import decimalai.deepagents
from deepagents import create_deep_agent

agent = create_deep_agent(model=model, **decimalai.deepagents.skills())
agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

The default backend reads files but has no shell. To run a skill's scripts, pass skills(backend=LocalShellBackend(...)).

The same install from Python, for a notebook or a setup script. Deep Agents' own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" deepagents

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("skydoves-migrating-to-modifier-node")
if result.exit_code:
    raise SystemExit(result.problems)

import decimalai.deepagents
from deepagents import create_deep_agent

agent = create_deep_agent(model=model, **decimalai.deepagents.skills())
agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

The default backend reads files but has no shell. To run a skill's scripts, pass skills(backend=LocalShellBackend(...)).

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/skydoves-migrating-to-modifier-node/SKILL.md

Give it to your agent

from pathlib import Path
from deepagents import create_deep_agent

skill = Path(".agents/skills/skydoves-migrating-to-modifier-node/SKILL.md").read_text()
agent = create_deep_agent(model=model, system_prompt=f"You are a helpful agent.\n\n{skill}")
agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Deep Agents docs

Its own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" agent-framework
decimalai skills add skydoves-migrating-to-modifier-node

Add it to your agent

import asyncio

from agent_framework import Agent
from decimalai.agent_framework import skills

agent = Agent(client=client, instructions="You are a helpful agent.", context_providers=[skills()])
asyncio.run(agent.run("Hello!"))

It caches skills by default, so restart the process after decimalai skills update.

The same install from Python, for a notebook or a setup script. Its own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" agent-framework

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("skydoves-migrating-to-modifier-node")
if result.exit_code:
    raise SystemExit(result.problems)

import asyncio

from agent_framework import Agent
from decimalai.agent_framework import skills

agent = Agent(client=client, instructions="You are a helpful agent.", context_providers=[skills()])
asyncio.run(agent.run("Hello!"))

It caches skills by default, so restart the process after decimalai skills update.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/skydoves-migrating-to-modifier-node/SKILL.md

Give it to your agent

import asyncio
from pathlib import Path
from agent_framework import Agent

skill = Path(".agents/skills/skydoves-migrating-to-modifier-node/SKILL.md").read_text()
agent = Agent(client=client, instructions=f"You are a helpful agent.\n\n{skill}")
asyncio.run(agent.run("Hello!"))

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Microsoft Agent Framework docs

Strands' own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" strands-agents
decimalai skills add skydoves-migrating-to-modifier-node

Add it to your agent

from decimalai.strands import skills
from strands import Agent
from strands.vended_tools import file_editor, shell

agent = Agent(
    model=model,                # always pass one: the default is Amazon Bedrock
    system_prompt="You are a helpful agent.",
    tools=[file_editor, shell],
    plugins=[skills()],
)
agent("Hello!")

Strands loads an agent's skills once, so build the agent again after decimalai skills update.

The same install from Python, for a notebook or a setup script. Strands' own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" strands-agents

Add it to your agent

from decimalai import skillpkg

# Writes .agents/skills and .decimal/skills.toml. Safe to run twice.
result = skillpkg.add("skydoves-migrating-to-modifier-node")
if result.exit_code:
    raise SystemExit(result.problems)

from decimalai.strands import skills
from strands import Agent
from strands.vended_tools import file_editor, shell

agent = Agent(
    model=model,                # always pass one: the default is Amazon Bedrock
    system_prompt="You are a helpful agent.",
    tools=[file_editor, shell],
    plugins=[skills()],
)
agent("Hello!")

Strands loads an agent's skills once, so build the agent again after decimalai skills update.

No DecimalAI package. The whole skill goes in every prompt, and nothing pins or updates it.

Copy SKILL.md

Save it in your project, as

.agents/skills/skydoves-migrating-to-modifier-node/SKILL.md

Give it to your agent

from pathlib import Path
from strands import Agent

skill = Path(".agents/skills/skydoves-migrating-to-modifier-node/SKILL.md").read_text()
agent = Agent(model=model, system_prompt=f"You are a helpful agent.\n\n{skill}")
agent("Hello!")

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Strands docs

Adds every file the skill has, pinned so you can update it. Cursor, Copilot, Cline and Warp read this folder by themselves.

pip install "decimalai>=0.15"
decimalai skills add skydoves-migrating-to-modifier-node

No Python or account. Cursor, Copilot, Cline and Warp read this folder by themselves.

Save it into your project

curl -fsSL --create-dirs -o .agents/skills/skydoves-migrating-to-modifier-node/SKILL.md https://app.decimal.ai/s/skydoves-migrating-to-modifier-node/SKILL.md

Copy SKILL.md

Save it in your project, as

.agents/skills/skydoves-migrating-to-modifier-node/SKILL.md

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Cursor and other editors docs

Fetch the file and give its text to your model as instructions. No account or package needed.

Get the file

curl -fsSL https://app.decimal.ai/s/skydoves-migrating-to-modifier-node/SKILL.md

Read the summary as JSON

curl -fsSL https://app.decimal.ai/s/skydoves-migrating-to-modifier-node.json

This gets SKILL.md only. The skill's 1 other file is not included.

Other ways to add it

Other setups docs

You don't need a DecimalAI account.

Version: v1, tested 8/19/2026
Source: skydoves/compose-performance-skills ↗
License: Apache-2.0

Manage it in DecimalAI

Saved prompts and skills, pinned versions, and traces of this skill in your runs. DecimalAI does not run your agent.

Claude Code reads files from your project, so DecimalAI config does not apply there.

On Claude Agent SDK: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On OpenAI Agents: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On OpenAI SDK: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On Anthropic: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On Gemini API: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On LangChain: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On CrewAI: set DECIMAL_API_KEY and add decimalai.init(crewai=True) to record runs and skill use. The skill version stays in your repo.

On Google ADK: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On Pydantic AI: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On Deep Agents: set DECIMAL_API_KEY and add decimalai.init(langchain=True) to record runs and skill use. The skill version stays in your repo.

On Microsoft Agent Framework: set DECIMAL_API_KEY and add decimalai.init(otel=True, agent_name="support") to record runs and skill use. The skill version stays in your repo.

On Strands: set DECIMAL_API_KEY and add decimalai.init(otel=True, agent_name="support") to record runs and skill use. The skill version stays in your repo.

Editors read files from your project, so DecimalAI config does not apply there.

Anywhere else: record a use with decimalai.log_skill_activation() inside a trace.

Use this skill

SKILL.md

Migrating to Modifier.Node — Persistent Nodes Over `composed { }`

Modifier.composed { } allocates a fresh composable scope per modifier per composition: it cannot be skipped, cannot be hoisted, and forces the parent to run on every recomposition. Modifier.Node is a persistent node diffed by ModifierNodeElement.equals() — created once on first apply, updated in place on subsequent applies. There is no per-recomposition allocation, no fresh composable scope, and no parent invalidation chain, which is why Android Developers describes the system as "designed from the ground up to be far more performant" than the legacy composed { } factory (see developer.android.com/develop/ui/compose/custom-modifiers). This skill teaches Claude how to author new modifiers as Modifier.Node and migrate legacy composed { } factories.

When to use this skill

A custom modifier currently uses Modifier.composed { } (search the module for Modifier.composed).
Authoring a new custom modifier from scratch — never start with composed { }.
A code review surfaces a composed { } factory.
The custom modifier needs a CoroutineScope (animation loop, debouncer), reads a CompositionLocal, participates in layout / drawing / pointer input, or tracks layout coordinates.
A @TraceRecomposition log shows the parent composable recomposing on every frame because a composed { } modifier is in the chain.

When NOT to use this skill

The "modifier" is actually a one-line composable wrapper that can stay a @Composable function — leave it alone.
The built-in modifier composition (Modifier.padding(...).clickable(...)) is sufficient, no custom node behavior needed.
The fix the developer needs is reordering an existing chain, not authoring a new node — see ../ordering-modifier-chains/SKILL.md.
The custom modifier reads a hot animation value via Modifier.composed { } only to feed a value-form modifier underneath — the underlying issue is a wrong-phase state read; see ../../recomposition/deferring-state-reads/SKILL.md.

Prerequisites

Compose UI 1.5+ — Modifier.Node and the specialized node interfaces are stable here.
Kotlin 2.0+ with org.jetbrains.kotlin.plugin.compose applied. Strong Skipping is on by default; non-skippable modifiers compound at scroll velocity.
A release build for any final measurement (skydoves hot take #5: debug builds run interpreted and lie).
For the full per-interface override surface and lifecycle diagram, read references/modifier-node-anatomy.md before authoring anything beyond a DrawModifierNode.

Workflow

1. Identify every Modifier.composed { } factory in the module. Grep for Modifier.composed. Each match is one migration target. Note what the body does — remember, drawBehind, LaunchedEffect, a CompositionLocal read, a pointer handler — because that decides which specialized node interface(s) you need.
2. Sketch the three pieces every Modifier.Node migration produces.
- (a) The public extension — fun Modifier.foo(...): Modifier = this then FooElement(...). Same name and signature as the old composed { } factory.
- (b) The ModifierNodeElement<T> data class — holds the parameters; implements create() (called once on first apply) and update(node: T) (called on subsequent applies).
- (c) The Modifier.Node subclass — holds mutable state (var fields), implements one or more specialized node interfaces, and runs lifecycle hooks (onAttach, onDetach, onReset).
3. Make the Element a data class. The compiler-synthesized equals()/hashCode() are how Compose decides whether to call update() vs leave the node alone. MUST be data class. A plain class falls back to referential equality, the diff thinks every apply is a new modifier, and update() is never called — your node holds stale parameters silently.
4. Pick the right specialized node interface(s). A Modifier.Node is empty by itself; behavior comes from interfaces it implements. The common ones:

Interface	Use when the modifier needs to …
`DrawModifierNode`	draw (replaces `drawBehind`/`drawWithCache`)
`LayoutModifierNode`	measure/place (replaces `layout { }` and custom `Layout`)
`SemanticsModifierNode`	contribute to accessibility
`PointerInputModifierNode`	handle pointer / gesture input (replaces `pointerInput`)
`CompositionLocalConsumerModifierNode`	read a `CompositionLocal` from inside the node
`LayoutAwareModifierNode`	get notified when this node's size/coordinates change
`GlobalPositionAwareModifierNode`	get notified about position in the window/root
`ObserverModifierNode`	observe arbitrary state reads with a custom observer (`observeReads { ... }`)
`DelegatingNode`	compose multiple node behaviors by delegating to child nodes
`TraversableNode`	walk the modifier chain (parent/child traversal)

A node MAY implement several interfaces at once — e.g. DrawModifierNode + CompositionLocalConsumerModifierNode + LayoutAwareModifierNode. For complex multi-behavior modifiers, PREFERRED: compose smaller DelegatingNode children rather than one mega-node implementing five interfaces.

5. Use the built-in coroutineScope for async work. Every Modifier.Node exposes a coroutineScope: CoroutineScope lazily tied to the node's attach/detach lifecycle. Launch animations, observers, debouncers there from onAttach(). MUST NOT create your own CoroutineScope inside onAttach — you will leak it past onDetach.
6. Trigger re-runs explicitly when needed. When you mutate node state from inside update() or a coroutine and need a redraw / re-measure / re-place, call invalidateDraw(), invalidateMeasurement(), or invalidatePlacement(). By default, update() triggers an auto-invalidation; for fine control, override shouldAutoInvalidate = false and invalidate manually.
7. Implement onAttach/onDetach/onReset for resource lifecycle. onAttach runs when the node joins the tree; onDetach when it leaves; onReset when the node is reused (only relevant inside lazy layouts). MUST release listeners, observers, and external subscriptions in onDetach.
8. Verify migration: no Modifier.composed { } remains. Re-grep the module. If any composed { } calls survive, list them with rationale; otherwise the migration is complete.

Patterns

Pattern: migrate `composed { drawBehind }` to `DrawModifierNode`

// WRONG (legacy)
fun Modifier.circle(color: Color): Modifier = composed {
    val computed = remember(color) { color.copy(alpha = 0.5f) }
    drawBehind { drawCircle(computed) }
}
// WRONG because: composed { } opens a fresh composable scope per parent recomposition; the modifier can never be skipped and forces the parent to recompose on every read it does inside.

// RIGHT
private data class CircleElement(val color: Color) : ModifierNodeElement<CircleNode>() {
    override fun create(): CircleNode = CircleNode(color)
    override fun update(node: CircleNode) { node.color = color }
}

private class CircleNode(var color: Color) : Modifier.Node(), DrawModifierNode {
    override fun ContentDrawScope.draw() {
        drawCircle(color.copy(alpha = 0.5f))
        drawContent()
    }
}

fun Modifier.circle(color: Color): Modifier = this then CircleElement(color)

The data class Element gives equals() / hashCode() for free. When the caller passes the same color, equals() returns true and the node is left alone. When the color changes, update() mutates node.color in place — no allocation, no Composition invalidation in the parent.

Pattern: coroutine work in a node (replace `composed { LaunchedEffect }`)

// WRONG (legacy)
fun Modifier.pulse(period: Long): Modifier = composed {
    val alpha = remember { Animatable(1f) }
    LaunchedEffect(period) {
        while (true) { alpha.animateTo(0.3f); alpha.animateTo(1f); delay(period) }
    }
    graphicsLayer { this.alpha = alpha.value }
}
// WRONG because: every parent recomposition allocates a new composable scope, and the LaunchedEffect's keying logic re-evaluates inside that scope.

// RIGHT
private data class PulseElement(val period: Long) : ModifierNodeElement<PulseNode>() {
    override fun create(): PulseNode = PulseNode(period)
    override fun update(node: PulseNode) { node.period = period }
}

private class PulseNode(var period: Long) : Modifier.Node(), DrawModifierNode {
    private var alpha by mutableFloatStateOf(1f)

    override fun onAttach() {
        coroutineScope.launch {
            while (true) {
                animate(1f, 0.3f) { value, _ -> alpha = value; invalidateDraw() }
                animate(0.3f, 1f) { value, _ -> alpha = value; invalidateDraw() }
                delay(period)
            }
        }
    }

    override fun ContentDrawScope.draw() {
        drawContent()
        drawRect(Color.Black.copy(alpha = 1f - alpha), blendMode = BlendMode.DstIn)
    }
}

fun Modifier.pulse(period: Long): Modifier = this then PulseElement(period)

The node's built-in coroutineScope is cancelled automatically on onDetach. No leak, no manual DisposableEffect.

Pattern: read a `CompositionLocal` inside a node

// WRONG (legacy)
fun Modifier.themedBorder(width: Dp): Modifier = composed {
    val tokens = LocalThemeTokens.current
    drawBehind { drawRect(tokens.outline, style = Stroke(width.toPx())) }
}
// WRONG because: every CompositionLocal read inside composed { } pins the modifier to a fresh scope per parent recomposition.

// RIGHT
private data class ThemedBorderElement(val width: Dp) : ModifierNodeElement<ThemedBorderNode>() {
    override fun create(): ThemedBorderNode = ThemedBorderNode(width)
    override fun update(node: ThemedBorderNode) { node.width = width }
}

private class ThemedBorderNode(
    var width: Dp,
) : Modifier.Node(), DrawModifierNode, CompositionLocalConsumerModifierNode {
    override fun ContentDrawScope.draw() {
        val tokens = currentValueOf(LocalThemeTokens)
        drawContent()
        drawRect(tokens.outline, style = Stroke(width.toPx()))
    }
}

fun Modifier.themedBorder(width: Dp): Modifier = this then ThemedBorderElement(width)

CompositionLocalConsumerModifierNode exposes currentValueOf(local) from inside any node callback. Reads are tracked by the Draw invalidation list, not by a Composition restart scope, so changing LocalThemeTokens redraws the node without recomposing the parent.

Pattern: forgetting `data class` (the silent-stale-node bug)

// WRONG
private class CircleElement(val color: Color) : ModifierNodeElement<CircleNode>() {
    override fun create() = CircleNode(color)
    override fun update(node: CircleNode) { node.color = color }
}
// WRONG because: not a data class -> equals() is referential -> every apply looks like a different element -> Compose tears down and recreates the node every time, OR the diff fails and update() is never called, leaving the node with the original color forever.

// RIGHT
private data class CircleElement(val color: Color) : ModifierNodeElement<CircleNode>() {
    override fun create() = CircleNode(color)
    override fun update(node: CircleNode) { node.color = color }
}

Pattern: holding a reference to the calling composable

// WRONG
private class HostingNode(val composer: Composer) : Modifier.Node() { /* ... */ }
// WRONG because: a Modifier.Node outlives any single composition pass; holding a Composer/composition-scoped object leaks it and invokes undefined behavior.

// RIGHT — accept primitive/stable parameters; read CompositionLocals via CompositionLocalConsumerModifierNode if you need composition context.
private data class HostingElement(val tag: String) : ModifierNodeElement<HostingNode>() {
    override fun create() = HostingNode(tag)
    override fun update(node: HostingNode) { node.tag = tag }
}
private class HostingNode(var tag: String) : Modifier.Node() { /* ... */ }

Pattern: composing multiple behaviors with `DelegatingNode`

// RIGHT — one public modifier, three small nodes delegated under one element
private data class CardEffectsElement(
    val color: Color,
    val onClick: () -> Unit,
) : ModifierNodeElement<CardEffectsNode>() {
    override fun create() = CardEffectsNode(color, onClick)
    override fun update(node: CardEffectsNode) {
        node.update(color, onClick)
    }
}

private class CardEffectsNode(
    color: Color,
    onClick: () -> Unit,
) : DelegatingNode() {
    private val background = delegate(BackgroundNode(color))
    private val click = delegate(ClickNode(onClick))

    fun update(color: Color, onClick: () -> Unit) {
        background.color = color
        click.onClick = onClick
    }
}

DelegatingNode is the canonical way to assemble multi-behavior modifiers without one node implementing every interface. The delegated children share the host's lifecycle.

Specialized node interfaces

Cheat sheet — full override surface and "use when" guidance lives in references/modifier-node-anatomy.md:

DrawModifierNode — implement ContentDrawScope.draw(). Replaces drawBehind/drawWithCache for custom modifiers.
LayoutModifierNode — implement MeasureScope.measure(...). Replaces Modifier.layout { }.
SemanticsModifierNode — implement SemanticsPropertyReceiver.applySemantics().
PointerInputModifierNode — implement onPointerEvent(...) and onCancelPointerInput().
CompositionLocalConsumerModifierNode — exposes currentValueOf(local) inside any node callback.
LayoutAwareModifierNode — onPlaced(coordinates) / onRemeasured(size).
GlobalPositionAwareModifierNode — onGloballyPositioned(coordinates).
ObserverModifierNode — wrap state reads with observeReads { ... } and react in onObservedReadsChanged().
DelegatingNode — delegate(otherNode) to compose behaviors.
TraversableNode — walk parents/children/descendants via the top-level extension functions on DelegatableNode: traverseAncestors(key, block), traverseChildren(key, block), traverseDescendants(key, block). The key parameter selects which traversable nodes participate; the descendants overload's block returns a TraverseDescendantsAction (continue / skip / cancel).

Lifecycle (short form — full diagram in references)

ModifierNodeElement.create()         // first apply only
        ↓
Modifier.Node.onAttach()             // node joins the tree; coroutineScope becomes valid
        ↓                            // (lives here across many parent recompositions)
ModifierNodeElement.update(node)     // each subsequent apply with !equals previous
        ↓                            // mutate node.var fields; auto-invalidates by default
Modifier.Node.onReset()              // optional: lazy-layout reuse
        ↓
Modifier.Node.onDetach()             // node leaves the tree; coroutineScope is cancelled

Mandatory rules

MUST prefer Modifier.Node for any new custom modifier — Modifier.composed { } is legacy.
MUST make ModifierNodeElement<T> a data class so the synthesized equals()/hashCode() drive the diff. Plain class silently breaks update().
MUST override update(node: T) to mutate node state in place. MUST NOT recreate the node from update().
MUST release subscriptions, listeners, and external resources in onDetach(). coroutineScope is cancelled for you; manual resources are not.
MUST NOT hold a reference to the Composer, the calling composable, the parent composition, or any composition-scoped object inside a Modifier.Node.
MUST NOT allocate a new CoroutineScope in onAttach — use the built-in coroutineScope property.
MUST NOT recommend Modifier.composed { } for new code. (Repo-wide rule from SPEC §8.)
PREFERRED: specialized node interfaces (DrawModifierNode, LayoutModifierNode, …) over a bare Modifier.Node.
PREFERRED: DelegatingNode over implementing more than ~3 specialized interfaces on a single node.
PREFERRED: override shouldAutoInvalidate = false and call invalidateDraw() / invalidateMeasurement() / invalidatePlacement() explicitly when fine-grained control matters; otherwise rely on the auto-invalidation triggered by update().

Verification

grep -R "Modifier.composed" <module>/src returns zero matches in the migrated module.
Every migrated modifier exposes a data class Element (search: class .*Element : ModifierNodeElement should be data class).
The compiler report (composables.txt) shows the parent composables that consume the migrated modifier are now restartable skippable (no composed-induced non-skip).
Layout Inspector → Recomposition Counts on the parent composable plateaus across animation frames driven by the modifier; only the node's draw/layout invalidation list ticks.
@TraceRecomposition (skydoves/compose-stability-analyzer) on the parent confirms in release + R8 + real device that the parent is not recomposed by the modifier's internal animation.
Resources allocated in onAttach (listeners, observers) are released in onDetach — verify with a leak canary pass.

References

Android Developers — Custom modifiers (Modifier.Node): https://developer.android.com/develop/ui/compose/custom-modifiers
Android Developers — Graphics modifiers: https://developer.android.com/develop/ui/compose/graphics/draw/modifiers
Android Developers — Performance overview: https://developer.android.com/develop/ui/compose/performance
Android Developers — Phases of Jetpack Compose: https://developer.android.com/develop/ui/compose/phases
Ben Trengrove — Debugging recomposition: https://medium.com/androiddevelopers/jetpack-compose-debugging-recomposition-bfcf4a6f8d37
Chris Banes — Compose performance tag: https://chrisbanes.me/tags/jetpack-compose-performance/
skydoves — 6 Jetpack Compose Guidelines: https://medium.com/proandroiddev/6-jetpack-compose-guidelines-to-optimize-your-app-performance-be18533721f9
skydoves — Jetpack Compose Mechanism (slides): https://speakerdeck.com/skydoves/jetpack-compose-mechanism
references/modifier-node-anatomy.md — full per-interface override surface, lifecycle diagram, DelegatingNode composition recipes, and manual-invalidation guidance.

About this skill

Automatic

Frontend / UI pack

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

skydoves / compose-performance-skills

modifiers/migrating-to-modifier-node/SKILL.md

View on GitHub ↗

License Apache-2.0. See LICENSE for complete terms.Imported 8/3/2026Version v1Bundled 1 file

Illustrative Want to see what skydoves/migrating-to-modifier-node would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Other skills for the same job

Published skills whose descriptions closely match this one, with their headline benchmark lift.

skydoves/ordering-modifier-chainsUse this skill to diagnose and fix Jetpack Compose Modifier ordering bugs — wrong paint region for background, wrong click area for clickable, wrong clipping for clip, wrong measurement for padding/size, surprising graphicsLayer scope. Covers the wrap-the-next-modifier mental model, the canonical pitfalls (padding vs background, clickable placement, clip before background, graphicsLayer placement), and why hoisting an entire Modifier chain via remember { Modifier.… } is rarely a real perf win beClose match-33 pts pass-rate lift

Related skills

Skills that do similar work, with their headline benchmark lift.

From the same source and nearby in the catalogue.

Ready to supercharge your AI agents?

SkillScore

Safety

What it changes

Track record

Use this skill

Migrating to Modifier.Node — Persistent Nodes Over composed { }

When to use this skill

When NOT to use this skill

Prerequisites

Workflow

Patterns

Pattern: migrate composed { drawBehind } to DrawModifierNode

Pattern: coroutine work in a node (replace composed { LaunchedEffect })

Pattern: read a CompositionLocal inside a node

Pattern: forgetting data class (the silent-stale-node bug)

Pattern: holding a reference to the calling composable

Pattern: composing multiple behaviors with DelegatingNode

Specialized node interfaces

Lifecycle (short form — full diagram in references)

Mandatory rules

Verification

References

SkillSafety

Source

Other skills for the same job

Related skills

More in the registry

Migrating to Modifier.Node — Persistent Nodes Over `composed { }`

Pattern: migrate `composed { drawBehind }` to `DrawModifierNode`

Pattern: coroutine work in a node (replace `composed { LaunchedEffect }`)

Pattern: read a `CompositionLocal` inside a node

Pattern: forgetting `data class` (the silent-stale-node bug)

Pattern: composing multiple behaviors with `DelegatingNode`