Ready to supercharge your AI agents?

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

Yellow Settlement Room · internet-court | DecimalAI

Financeby internet-court

Yellow Settlement Room

Yellow Network Protocol app sessions for AI agents - a shared room where several agents pool funds, reallocate off-chain at machine speed, and settle one final split.

Never made a case worseMaintained

SkillScore

SkillScore86/100

Test results 86

Safety

3/3 safety checks passed

Code scan
Intent
Content

What it changes

18%86%of test cases passed +68 pts

Without skill4 of 22

With skill19 of 22

15 fixed4 passed either way3 still failing0 broke

Measured on gemini-3.6-flash9/26/2026v1

Track record

Agent runs: 0; All time

See one task, with and without the skill

One test taskWhen building state updates using decimal arithmetic for asset amounts in TypeScript under NodeNext module resolution, I am using import Decimal from 'decimal.js'. Will this compile and instantiate correctly?

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 internet-court-yellow-settlement-room
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("internet-court-yellow-settlement-room")
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 internet-court-yellow-settlement-room

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/internet-court-yellow-settlement-room/SKILL.md https://app.decimal.ai/s/internet-court-yellow-settlement-room/SKILL.md

Copy SKILL.md

Save it in your project, as

.claude/skills/internet-court-yellow-settlement-room/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 internet-court-yellow-settlement-room

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("internet-court-yellow-settlement-room")
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/internet-court-yellow-settlement-room/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/internet-court-yellow-settlement-room/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 internet-court-yellow-settlement-room

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("internet-court-yellow-settlement-room")
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/internet-court-yellow-settlement-room/SKILL.md

Give it to your agent

from pathlib import Path
from agents import Agent, Runner

skill = Path(".agents/skills/internet-court-yellow-settlement-room/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 internet-court-yellow-settlement-room

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("internet-court-yellow-settlement-room")
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/internet-court-yellow-settlement-room/SKILL.md

Give it to your agent

from pathlib import Path
from openai import OpenAI

client = OpenAI()
skill = Path(".agents/skills/internet-court-yellow-settlement-room/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 internet-court-yellow-settlement-room

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("internet-court-yellow-settlement-room")
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/internet-court-yellow-settlement-room/SKILL.md

Give it to your agent

from pathlib import Path
import anthropic

client = anthropic.Anthropic()
skill = Path(".agents/skills/internet-court-yellow-settlement-room/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 internet-court-yellow-settlement-room

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("internet-court-yellow-settlement-room")
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/internet-court-yellow-settlement-room/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/internet-court-yellow-settlement-room/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 internet-court-yellow-settlement-room

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("internet-court-yellow-settlement-room")
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/internet-court-yellow-settlement-room/SKILL.md

Give it to your agent

from pathlib import Path
from langchain.agents import create_agent

skill = Path(".agents/skills/internet-court-yellow-settlement-room/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 internet-court-yellow-settlement-room

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("internet-court-yellow-settlement-room")
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/internet-court-yellow-settlement-room/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 internet-court-yellow-settlement-room

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("internet-court-yellow-settlement-room")
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/internet-court-yellow-settlement-room/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/internet-court-yellow-settlement-room/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 internet-court-yellow-settlement-room

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("internet-court-yellow-settlement-room")
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/internet-court-yellow-settlement-room/SKILL.md

Give it to your agent

from pathlib import Path
from pydantic_ai import Agent

skill = Path(".agents/skills/internet-court-yellow-settlement-room/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 internet-court-yellow-settlement-room

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("internet-court-yellow-settlement-room")
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/internet-court-yellow-settlement-room/SKILL.md

Give it to your agent

from pathlib import Path
from deepagents import create_deep_agent

skill = Path(".agents/skills/internet-court-yellow-settlement-room/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 internet-court-yellow-settlement-room

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("internet-court-yellow-settlement-room")
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/internet-court-yellow-settlement-room/SKILL.md

Give it to your agent

import asyncio
from pathlib import Path
from agent_framework import Agent

skill = Path(".agents/skills/internet-court-yellow-settlement-room/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 internet-court-yellow-settlement-room

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("internet-court-yellow-settlement-room")
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/internet-court-yellow-settlement-room/SKILL.md

Give it to your agent

from pathlib import Path
from strands import Agent

skill = Path(".agents/skills/internet-court-yellow-settlement-room/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 internet-court-yellow-settlement-room

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/internet-court-yellow-settlement-room/SKILL.md https://app.decimal.ai/s/internet-court-yellow-settlement-room/SKILL.md

Copy SKILL.md

Save it in your project, as

.agents/skills/internet-court-yellow-settlement-room/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/internet-court-yellow-settlement-room/SKILL.md

Read the summary as JSON

curl -fsSL https://app.decimal.ai/s/internet-court-yellow-settlement-room.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 9/26/2026
Source: internet-court/internet-court-skill ↗
License: NOASSERTION

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

Yellow Settlement Room

Use this skill when several AI agents need to hold funds together and settle one outcome: open a shared session, each agent's balance is tracked inside it, they reallocate between themselves off-chain with no gas per step, and they co-sign the final split.

The session holds N agents, not two. A payment rail moves value from one payer to one payee; a swarm of agents settling over a rail needs a separate escrow per pair. One session settles all of them at once: one object, one deposit per agent, one final allocation. That is the shape to reach for when more than two agents have a stake in the same outcome.

This skill covers the app-session (virtual) layer only. It operates on funds that are already in an account balance at Yellow. A session opens with zero allocations, so not every participant needs funds - only a participant that makes a deposit does. Getting funds into an account is a one-time on-chain step, out of scope here - see ## Prerequisite. This skill never funds accounts; before a participant deposits, check its balance with client.getBalances(wallet), and if it is short, stop and report it.

Package: @yellow-org/sdk (v1). A complete runnable reference is the official example at github.com/layer-3/docs, examples/nitrolite-v1-lifecycle; the flow below matches it exactly, generalised from two participants to N. For method lookups, the docs MCP: npx -y @yellow-org/sdk-mcp@^1.

Core Model

Each agent runs its own client with its own key (backend: private-key based)
  -> agents open one app session: N participants, signature weights, quorum
     (opens with ZERO allocations; nobody needs funds yet)
  -> a depositing agent commits its OWN funds into the session
     (only depositors need a funded account; others can hold zero)
  -> agents reallocate between themselves (operate), each update co-signed to quorum
  -> withdraw / close: the final split releases back to channels, withdrawable on-chain

The session is an off-chain ledger hosted by the Yellow node. Its guarantee is signature-based: no agent's allocation changes without signatures meeting the quorum. It is not a trustless escrow; read ## Trust Boundary before sizing exposure.

Design Rules

One agent, one key, one client. Each agent signs only with its own key, from its own process. A backend service is private-key based; a frontend is wallet based. You cannot take several agents' private keys into one client. Holding multiple participants' keys in one process is valid only as a local test or an explicitly custodial server. See ## Roles and key separation.
Never fund accounts from this skill. Funding happens once, on-chain, before a session. Only a participant that deposits needs funds; check its balance first with client.getBalances(wallet) and if it is short, stop and report the shortfall. Do not attempt deposit, approveToken, or transfer to self-fund.
Never call an allocation "locked on-chain and enforceable." Funds are committed out of a channel and governed by quorum, so no counterparty agent can take them - but releasing them still needs the node to co-sign. State both halves.
Default to equal weights and unanimous quorum. Only give a subset of agents combined weight >= quorum when that subset is intentionally trusted. See ## Weights and Quorum.

Prerequisite: a funded account for each depositor

A session is created with zero allocations, so not every participant needs funds. Only a participant that makes a deposit needs a funded account balance at Yellow for the asset. (An account is backed by an on-chain state channel, but you can treat it as the participant's balance.) That balance is the ceiling on what a depositor can commit and the most it can lose. Check a depositor's balance before it deposits:

const balances = await client.getBalances(wallet);   // account balances at Yellow, per asset

Funding is a one-time on-chain operation, done once before any session and out of scope here. The funding calls, for reference, are:

await client.setHomeBlockchain(asset, chainId);
await client.approveToken(chainId, asset, amount);
await client.deposit(chainId, asset, amount);
await client.checkpoint(asset);              // finalizes the deposit on-chain

For the exact funding sequence and any Node-specific transaction setup, see the quickstart at docs.yellow.org/nitrolite/build/getting-started/quickstart and the official nitrolite-v1-lifecycle example.

Connect

import { Client, createSigners, withBlockchainRPC } from '@yellow-org/sdk';

const signers = createSigners(privateKey);       // 0x-prefixed 32-byte hex
const client = await Client.create(
  wsURL,                                          // sandbox: wss://nitronode-sandbox.yellow.org/v1/ws
  signers.stateSigner, signers.txSigner,
  withBlockchainRPC(chainId, rpcURL),
);

There is no login handshake; authorization is per-call, from the signatures inside each payload. Each agent runs its own client with its own key, in its own process. The examples below show several signers together for readability; in a real agent-to-agent deployment each agent constructs only its own signer and signs the shared state hash with its own key.

Roles and key separation

Agent-to-agent means the keys are distributed. You cannot take several agents' private keys into one client. Model these roles:

Each agent holds its own key and runs its own client. A backend agent is private-key based; a frontend agent is wallet based. Every signer signs the same full-state hash independently with its own key, in its own process (this is independent co-signing, not partial or threshold signing).
Proposer (an agent, or a coordinating server): builds a state update, packs the hash, and sends that hash to each required signer.
Signers (the agents whose combined weight must meet quorum): each signs the packed hash with its own key and returns the signature.
Submitter (usually the proposer): collects the signatures and calls the Nitronode (createAppSession / submitAppSessionDeposit / submitAppState).

The cross-process pattern uses the same methods as the single-process code below:

// Proposer (any agent or a coordinating server): build the state, pack the hash.
const hash = packAppStateUpdateV1(update);   // portable 0x string, safe to send over the wire

// Each signer, in its OWN process with its OWN key:
const mySig = await new AppSessionWalletSignerV1(new EthereumMsgSigner(myKey)).signMessage(hash);
// ...return mySig to the proposer over your own transport (HTTP, queue, etc.)

// Submitter: gather signatures until summed weight meets quorum, then submit once.
await client.submitAppState(update, [sigFromA, sigFromB, sigFromC]);

The protocol carries no transport for moving the hash out and the signatures back; that is the integrator's to build. A common topology: a Nitronode, agents connecting to a coordinating server (an app), and agents transacting agent-to-agent through shared sessions. The single-process code below co-locates keys only for readability and local testing; do not ship it that way.

Open the session

import {
  AppSessionWalletSignerV1, EthereumMsgSigner,
  packCreateAppSessionRequestV1, type AppDefinitionV1,
} from '@yellow-org/sdk';

// One session signer per participant, a plain wallet signer (type 0xa1).
// LOCAL TEST ONLY: holding pkA, pkB, pkC in one process is a shortcut for a
// smoke test or a custodial server. In a real deployment each agent builds ONLY
// its own signer, in its own process, from its own key (see Roles and key
// separation above). Session keys are an optional friction-reducer, omitted here.
const signerA = new AppSessionWalletSignerV1(new EthereumMsgSigner(pkA));
const signerB = new AppSessionWalletSignerV1(new EthereumMsgSigner(pkB));
const signerC = new AppSessionWalletSignerV1(new EthereumMsgSigner(pkC));

const definition: AppDefinitionV1 = {
  applicationId: appId,                           // ^[a-z0-9_-]{1,66}$
  participants: [
    { walletAddress: addrA, signatureWeight: 1 },
    { walletAddress: addrB, signatureWeight: 1 },
    { walletAddress: addrC, signatureWeight: 1 },
  ],
  quorum: 3,                                       // = sum of weights -> unanimous (the safe default)
  nonce: BigInt(Date.now()) * 1_000_000n + BigInt(Math.floor(Math.random() * 1_000_000)),
};

const createPayload = packCreateAppSessionRequestV1(definition, sessionData);
const created = await client.createAppSession(definition, sessionData, [
  await signerA.signMessage(createPayload),
  await signerB.signMessage(createPayload),
  await signerC.signMessage(createPayload),       // creation must itself meet quorum
]);
// created.appSessionId, created.version, created.status

The participant set is immutable after creation; no agent can be added later. Creation must meet quorum, so every participant that makes up the quorum co-signs the create request.

Read the live state

const { sessions } = await client.getAppSessions({ appSessionId });
const session = sessions[0];    // session.version, session.isClosed, session.allocations

Read this immediately before signing any update: version must be exactly session.version + 1n.

Deposit, operate, withdraw, close

All updates share one shape; intent is a number, not a string.

import { AppStateUpdateIntent, packAppStateUpdateV1, type AppStateUpdateV1 } from '@yellow-org/sdk';
import { Decimal } from 'decimal.js';   // named import: default import is not constructable under NodeNext

// DEPOSIT (own endpoint): a depositor commits its OWN funds into the session.
// List ONLY the depositing participant in allocations; do not add zero-value
// entries for participants who are not depositing here.
const deposit: AppStateUpdateV1 = {
  appSessionId, intent: AppStateUpdateIntent.Deposit, version: session.version + 1n,
  allocations: [
    { participant: addrA, asset, amount: new Decimal('10') },
  ],
  sessionData: JSON.stringify({ intent: 'fund' }),
};
const dp = packAppStateUpdateV1(deposit);
// The deposit state still needs signatures meeting quorum. Each agent signs the
// hash with its OWN key in its OWN process; here they are shown together only for
// readability. The submitter gathers the signatures and calls the node.
await client.submitAppSessionDeposit(
  deposit, [await signerA.signMessage(dp), await signerB.signMessage(dp), await signerC.signMessage(dp)],
  asset, new Decimal('10'),                        // amount must equal the deposit allocation total
);

// OPERATE: reallocate between participants. Per-asset totals must stay CONSTANT,
// and every non-zero allocation must be restated (not a delta).
const operate: AppStateUpdateV1 = {
  appSessionId, intent: AppStateUpdateIntent.Operate, version: /* live */ session.version + 1n,
  allocations: [
    { participant: addrA, asset, amount: new Decimal('4') },
    { participant: addrB, asset, amount: new Decimal('4') },
    { participant: addrC, asset, amount: new Decimal('2') },
  ],
  sessionData: JSON.stringify({ round: 'payout' }),
};
const op = packAppStateUpdateV1(operate);
await client.submitAppState(operate, [ /* signatures summing to quorum */
  await signerA.signMessage(op), await signerB.signMessage(op), await signerC.signMessage(op),
]);

Withdraw (intent 2) may only decrease allocations and releases to channels. Close (intent 3) must restate the current allocation exactly, releases everything, and is terminal - never close while work or a review is outstanding. Both use submitAppState the same way.

Signatures are collected across agents, off the wire. The proposer builds the state and hash; each agent signs that hash with its own key in its own process; the submitter gathers the signatures until summed weight meets quorum and calls the node. The protocol provides no transport for this exchange - it is the caller's responsibility. Duplicate signers count once.

Weights and Quorum

signatureWeight per participant, quorum = the weight threshold a state needs to be valid.

Equal weights, quorum = sum -> unanimous. The safe default. Use it unless you have a specific reason not to.
A subset of agents whose weights sum to >= quorum can settle without the others. That is a feature for a trusted operator (e.g. a client that funds and controls a swarm) and a hazard between adversaries: in a client/orchestrator/worker room at 1/1/1 quorum 2, the orchestrator and worker together can sign a split that pays the worker and zeroes the client. A single scalar quorum cannot protect every party's balance at N > 2. If the parties mutually distrust, give each depositor a blocking stake, or keep them in separate sessions.

State which regime a session is in when you design it.

Trust Boundary

State this before any agent puts value at risk. Do not soften it.

No agent in the session can take another's allocation. Every change needs signatures meeting quorum. This is the real, strong guarantee.
The Yellow node is trusted for liveness and honest relay. Sessions are off-chain; funds become enforceable on-chain only once released back to a channel as a node-co-signed state. If the node will not co-sign, there is no on-chain path out of the session. Yellow's own docs state the protocol is not fully trust-minimized. This is weaker than an on-chain escrow and stronger than a spending-allowance delegation (where the payer can drain the account at any time). Say which you rely on.
A session has no dispute mechanism, challenge, or timeout of its own. If quorum is never reached, funds stay in the session. Cooperative settlement is the supported path today.
The deposit is the exposure ceiling. A participant can lose at most what it committed into the session.

Composes with

GenLayer for subjective disputes: a session cannot judge whether work was acceptable; a jury can. The intended composition is verdict-informed cooperative settlement (parties agree up front to sign the allocation a verdict dictates). Adjudicator-enforced settlement against a non-cooperative party is roadmap, not shipped - do not place already-contested funds in a session.
Alkahest (vendored/arkhai/alkahest-user) for a bilateral, one-shot deal that needs trustless on-chain escrow with a reclaim timeout. Reach for a settlement room when the deal is multiparty and many-update, which a single bilateral escrow cannot express.

Failure Cases

Version race: version must be exactly session.version + 1n. Concurrent signers collide and one update fails cleanly (the node serializes). Re-read the live version, re-collect signatures, retry. Most common friction in a busy room.
Quorum never reached: permanent deadlock, no timeout. Prevent by design (cooperative weights); for adversarial bilateral deals use Alkahest.
Unfunded participant: submitAppSessionDeposit fails if the account is not funded (on the sandbox the node error reads no channel state to advance). This is the prerequisite, not a bug - check getBalances(wallet) first and report a shortfall.
Operate that drops a non-zero allocation or whose per-asset total drifts: rejected.
Close while work or a review is outstanding: terminal and unrecoverable.
Node fails to co-sign a release: surface as a stuck session; do not retry silently or report success.

Output Checklist

Which agents are in the session, and whether they cooperate or mutually distrust.
Participant set with weights and quorum, and the arithmetic showing no coalition can rob a party.
Each depositor's funded balance confirmed via getBalances(wallet), and its deposit as the max it can lose. Non-depositing participants need no funds.
The happy path: open, deposit, operate, (withdraw), close - and the disagreement path, or an explicit statement that the room deadlocks.
The trust disclosure from ## Trust Boundary, in plain language, before any value moves.

References

references/agent-lifecycle.md - the full multiparty flow as runnable code, matching the official nitrolite-v1-lifecycle example.

When it activates

Automatically, when a task matches. Use for multiparty settlement among agents. Covers connecting, the funded-account prerequisite, creating a session with participants + weights + quorum, deposit, operate, withdraw, close, and the trust boundary. Grounded in the official @yellow-org/sdk lifecycle example.

About this skill

Automatic

Coding pack

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v5 — 8/24/2026

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

internet-court / internet-court-skill

vendored/yellow/yellow-settlement-room/SKILL.md

View on GitHub ↗

License NOASSERTIONImported 8/24/2026Version v1Bundled 1 file

Illustrative Want to see what internet-court/yellow-settlement-room would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

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

Yellow Settlement Room

Core Model

Design Rules

Prerequisite: a funded account for each depositor

Connect

Roles and key separation

Open the session

Read the live state

Deposit, operate, withdraw, close

Weights and Quorum

Trust Boundary

Composes with

Failure Cases

Output Checklist

References

When it activates

SkillSafety

Source

Related skills

More in the registry