Ready to supercharge your AI agents?

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

Install and run NocoBase 3 with app-installer | DecimalAI

DevOpsby nocobase

Install and run NocoBase 3 with app-installer

Deploy a built NocoBase 3 application locally or over SSH to an offline Linux server, for production, with `@nocobase/app-installer` from the deployment archive `pnpm build --tar` writes, then upgrade, roll back and check it.

Never made a case worseMaintained

SkillScore

SkillScore86/100

Test results 86

Safety

3/3 safety checks passed

Code scan
Intent
Content

What it changes

14%86%of test cases passed +73 pts

Without skill3 of 22

With skill19 of 22

16 fixed3 passed either way3 still failing0 broke

Measured on gemini-3.8-flash11/2/2026v1

Track record

Agent runs: 0; All time

See one task, with and without the skill

One test taskI want to run the installer command using npx. Should I run `npx @nocobase/app-installer latest`, omit the version tag, or what tag 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 nocobase-nocobase-app-installer
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("nocobase-nocobase-app-installer")
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 nocobase-nocobase-app-installer

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/nocobase-nocobase-app-installer/SKILL.md https://app.decimal.ai/s/nocobase-nocobase-app-installer/SKILL.md

Copy SKILL.md

Save it in your project, as

.claude/skills/nocobase-nocobase-app-installer/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 nocobase-nocobase-app-installer

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("nocobase-nocobase-app-installer")
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/nocobase-nocobase-app-installer/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/nocobase-nocobase-app-installer/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 nocobase-nocobase-app-installer

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("nocobase-nocobase-app-installer")
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/nocobase-nocobase-app-installer/SKILL.md

Give it to your agent

from pathlib import Path
from agents import Agent, Runner

skill = Path(".agents/skills/nocobase-nocobase-app-installer/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 nocobase-nocobase-app-installer

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("nocobase-nocobase-app-installer")
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/nocobase-nocobase-app-installer/SKILL.md

Give it to your agent

from pathlib import Path
from openai import OpenAI

client = OpenAI()
skill = Path(".agents/skills/nocobase-nocobase-app-installer/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 nocobase-nocobase-app-installer

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("nocobase-nocobase-app-installer")
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/nocobase-nocobase-app-installer/SKILL.md

Give it to your agent

from pathlib import Path
import anthropic

client = anthropic.Anthropic()
skill = Path(".agents/skills/nocobase-nocobase-app-installer/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 nocobase-nocobase-app-installer

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("nocobase-nocobase-app-installer")
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/nocobase-nocobase-app-installer/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/nocobase-nocobase-app-installer/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 nocobase-nocobase-app-installer

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("nocobase-nocobase-app-installer")
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/nocobase-nocobase-app-installer/SKILL.md

Give it to your agent

from pathlib import Path
from langchain.agents import create_agent

skill = Path(".agents/skills/nocobase-nocobase-app-installer/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 nocobase-nocobase-app-installer

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("nocobase-nocobase-app-installer")
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/nocobase-nocobase-app-installer/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 nocobase-nocobase-app-installer

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("nocobase-nocobase-app-installer")
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/nocobase-nocobase-app-installer/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/nocobase-nocobase-app-installer/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 nocobase-nocobase-app-installer

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("nocobase-nocobase-app-installer")
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/nocobase-nocobase-app-installer/SKILL.md

Give it to your agent

from pathlib import Path
from pydantic_ai import Agent

skill = Path(".agents/skills/nocobase-nocobase-app-installer/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 nocobase-nocobase-app-installer

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("nocobase-nocobase-app-installer")
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/nocobase-nocobase-app-installer/SKILL.md

Give it to your agent

from pathlib import Path
from deepagents import create_deep_agent

skill = Path(".agents/skills/nocobase-nocobase-app-installer/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 nocobase-nocobase-app-installer

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("nocobase-nocobase-app-installer")
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/nocobase-nocobase-app-installer/SKILL.md

Give it to your agent

import asyncio
from pathlib import Path
from agent_framework import Agent

skill = Path(".agents/skills/nocobase-nocobase-app-installer/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 nocobase-nocobase-app-installer

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("nocobase-nocobase-app-installer")
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/nocobase-nocobase-app-installer/SKILL.md

Give it to your agent

from pathlib import Path
from strands import Agent

skill = Path(".agents/skills/nocobase-nocobase-app-installer/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 nocobase-nocobase-app-installer

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/nocobase-nocobase-app-installer/SKILL.md https://app.decimal.ai/s/nocobase-nocobase-app-installer/SKILL.md

Copy SKILL.md

Save it in your project, as

.agents/skills/nocobase-nocobase-app-installer/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/nocobase-nocobase-app-installer/SKILL.md

Read the summary as JSON

curl -fsSL https://app.decimal.ai/s/nocobase-nocobase-app-installer.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 11/2/2026
Source: nocobase/nocobase ↗
License: Not recorded

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

Install and run NocoBase 3 with app-installer

This Skill installs NocoBase that is already built — nothing on the machine is developed, only run. @nocobase/app-installer puts each release under releases/<version>_<build time>/app and runs it under pm2; config.yml, app.env, storage/, backups/ and logs/ sit beside the releases and survive every upgrade. It installs from --archive, a deployment archive built in an application project. This Skill decides when to use it and how to act on what it reports; --help is the reference for every flag.

Run it

Run the installer on the machine where the Agent works, with the registry named and --json. For SSH deployment this is the local machine, not the server:

npx --yes --registry="${NOCOBASE_REGISTRY:-https://registry.npmjs.org}" @nocobase/app-installer@beta <command> … --json

Use @beta for the published installer; do not fall back to an untagged invocation or @latest. Preserve an explicit version required for an existing installation or requested by the user, and use the snapshot tag when testing an unreleased local registry.

NocoBase 3 publishes to the public npm registry. NOCOBASE_REGISTRY is set only when someone points the shell at another registry, such as an unreleased snapshot; the installer records the registry in installer.json and suggests commands from it. The --yes before the package is npx's own: it downloads the installer without asking. The installer's --yes, after the command, is a separate answer that only the user gives.

Use the published beta channel unless the user specifies an exact version. An unreleased snapshot registry tags its packages latest; use that tag only when testing such a registry. Check --help before deploying to confirm the selected version supports deploy and --ssh.

--json prints one document on stdout and progress on stderr, in the same envelope as pnpm nocobase … --json. Read the document, not the exit code alone: ok, status (success, success-noop or failure), result, and on failure error.code, error.message, error.suggestions and error.details. Each suggestion has a message and, where there is one, a run of { command, args }: an executable and its arguments, run as given without a shell. app-installer 0.1.0-beta.1 and earlier, which a suggested command pinned to that version still runs, print status error and run as one shell line that runs as printed, so tell success from failure by ok.

Exit	Meaning
`0`	Done, or nothing to do (`success-noop`).
`1`	Failed; an application that was already running was not touched.
`2`	Invalid usage, a failed precheck, or a confirmation still needed; nothing was written.
`3`	An upgrade failed after switching and was rolled back; the previous release is running.
`4`	Rolling back failed too; the application is down.

Choose the route

A project is source code someone changes, created by nocobase-create-app and developed locally; an installation is a built NocoBase that only runs, made by this Skill, usually in production. Settle the route before installing anything:

The user asks to	Route
Install, create or try NocoBase, to develop with	The `nocobase-create-app` Skill
Deploy an application to production or a server	This Skill, `--archive`, from the application's deployment archive
Run it with Docker	The application's `nocobase-deployment` Skill and Docker configuration

"Install NocoBase on this server" with no project and no archive is not settled yet. Running the user's own application means creating it locally with nocobase-create-app, building its archive, and installing that here with --archive. Do not create a project on the server and run it with pnpm dev: that is development mode, not a deployment.
An installer.json whose source.kind is template was installed from the published Hub template by an earlier app-installer. The current one refuses it with STATE_UNSUPPORTED; leave it to the app-installer version that installed it.
When the working directory already holds installer.json, the installation exists: start with status.

Choose where the installer runs

When the user supplies an SSH server, keep the Agent, this Skill and the installer on the local machine. Read Remote deployment before proceeding: it covers local orchestration, offline execution and recovery. The server needs no Agent, Skill or npm access. The install, status, upgrade and rollback guidance below still defines the lifecycle; run remote operations with the same SSH options and absolute remote --dir.

When the Agent is already on the application server, use the local commands below. An archive is built where the sources are, not in the production installation. If the source project is available locally, follow its AGENTS.md and nocobase-deployment Skill to build; otherwise obtain the archive from the user or CI. Do not build or run a development project on the production server.

Build the archive

In the application project, whose own AGENTS.md and nocobase-deployment Skill are the authority on building:

APP_BASE_PATH=/crm pnpm build --target linux-x64 --node-version 24 --tar

--target and --node-version describe the machine that will run it, not the one building it: node -p "process.platform + '-' + process.arch" and node --version there; Alpine takes a -musl target. The installer refuses an archive built for another machine and names the command that fits.
APP_BASE_PATH is compiled into the client, /main when unset. The application is served there, and every later archive must be built for the same one.
A database other than SQLite needs its driver in the project before the build, such as pnpm add @nocobase/db-postgres@beta: an archive carries only the drivers it was built with.

The archive is storage/exports/dist.tar.gz; the user copies it to the machine that runs NocoBase, for example scp storage/exports/dist.tar.gz user@server:/tmp/crm.tar.gz. ARCHIVE_TOO_OLD or STORAGE_IN_RELEASE means the project's NocoBase packages are older than the installer needs: upgrade them, build again, then retry.

Install

Every installation

Check the machine: node --version (24 or later) and command -v pm2, which finds pm2 without starting its daemon. pm2 4.3 or later has to be installed globally, npm install -g pm2; a copy fetched through npx breaks pm2 startup. Report what is missing with the command that installs it, and install nothing globally unless the user asks. On Windows, work in WSL.
Settle with the user: the target directory, new or empty, such as /srv/nocobase/crm; the public origin without the base path, such as https://apps.example.com, which may be left out for a machine reached only at http://127.0.0.1:<port>; whether a reverse proxy sits in front; the port, 13000 by default and different for every installation on the machine; and the database, SQLite by default.
It listens on 127.0.0.1, which suits a reverse proxy on the same machine; when people reach it directly at http://<address>:<port>, add --host 0.0.0.0 and use that address as the origin. For another database add --dialect <dialect>, the connection as --set database.connections.main.host=… and friends, and the password as --set-from-env database.connections.main.password=<VARIABLE> after the user has exported it; never put a password on the command line.
PORT_IN_USE names a free port in error.details.freePort; offer it rather than picking one silently.

Run it

install <dir> --archive <file> --origin <origin> --json. It needs no pnpm, and finishes in a minute or two. The archive must carry the driver for --dialect, or the install stops with DRIVER_MISSING and the build command that fixes it.

When it succeeds

When result.started is true, the installation answered its health check. Tell the user:

result.url, and the first sign-in: result.initialAdmin.username or its email, with the password under users.initialAdmin in config.yml. When result.initialAdmin.defaultPassword is true, that password is still the admin123 from config.example.yml and has to be changed after signing in; otherwise it is the one they set. The result never carries the password, and config.yml stays unread;
each command in result.nextCommands: pm2 startup prints a command they run once with sudo, after which pm2 restores the process list the installer saved, so the installation comes back after a reboot;
with a reverse proxy: it forwards to the installation's port with the WebSocket upgrade headers. Several installations behind one origin are routed by their base paths;
anything in warnings.

A failed install before the switch leaves the target as it found it. What the failing step said is in error.details: output for unpacking the release, stderr and stdout when the release's own CLI printed no result; a failure that CLI reports itself keeps the CLI's own code, such as CONFIG_INVALID, and its error.message and error.suggestions say what was wrong. START_FAILED means installed but not running: show error.details.log.

Several installations on one machine

Each application is installed once, in a directory of its own, with its own --port. The pm2 process name defaults to nocobase-<directory name>, so directories with different names never collide; two directories with the same name report PM2_NAME_IN_USE before anything is written, and --name gives the second its own.

Status

status --dir <root> --json changes nothing. Report the application and version, current (the release id), endpoints.url, health.ok, process, and anything in warnings. node.matches: false means the machine's Node major changed since the release was built, and it will not load its native modules until a release built for this machine replaces it: the warning names the steps — a rebuilt archive and upgrade --archive — which go through the same stop, backup and confirmation as any upgrade. A non-null pending means an operation was interrupted; see below.

Upgrade and roll back

Before upgrading an existing database, read the target application's and registered plugins' release-specific upgrade notes. Include documented checksum changes, data prerequisites, and irreversible migrations in the recovery plan. Before the installer stops the application, inspect the target application's db repair --dry-run --json output using its supported CLI and deployment configuration, even when no historical correction is documented. This is a read-only checksum check: the installer's own db apply --dry-run discovers pending tasks but does not validate checksums, even under a strict checksum policy. If the selected installer cannot perform the required check or preparation, establish the supported maintenance procedure before proceeding with the upgrade.

When the checksum preview proposes changes, explain every migration or seed checksum difference before any actual repair; when it proposes none, skip repair. The installer applies migrations but does not automatically acknowledge changed historical checksums. Repair only updates history checksums; it does not change schema, fix partial tables, or replace a forward migration. Do not suppress drift by changing migration policy or repeatedly retrying the same archive.

Do not run actual repair before an ordinary installer upgrade. Its SQLite backup is taken after stopping the application and would already contain the new checksums, which the previous runtime can reject under onChecksumMismatch: error if the upgrade rolls back. A repair procedure must stop writers, back up the original history, perform the reviewed repair and migrations, and restore that same pre-repair backup before restarting the old runtime on failure. An extra manual backup alone does not change which backup automatic rollback restores. When the selected installer cannot coordinate that sequence, finish only the read-only checks and establish the supported maintenance procedure first.

Both stop the application, so the user decides:

An upgrade takes a new archive: upgrade --dir <root> --archive <file> --json. It must be the same application built for the same base path, and not an older version; the same version built again is a new release and deploys normally.
Run it without --yes. It answers CONFIRMATION_REQUIRED with error.details.notes: what will stop, what is backed up, and any Node major change. Relay those notes and wait for a clear yes before running it again with --yes. success-noop means that release is already running.
The installer backs up every SQLite database config.yml declares, plus config.yml and app.env, but not any other database: BACKUP_REQUIRED names those connections; have the user back them up first, then add --backup-done.
An upgrade keeps three releases, always including the new one and the one it came from, and prunes older ones once it succeeds; pass --keep <n> when the user wants more of them to stay available for rollback.
Read the outcome:
- Exit 0: the new release passed its health check. Report result.fromVersion, result.toVersion, the release id result.to, result.migrations and result.backup, and name any release in result.pruned, which can no longer be rolled back to.
- Exit 3: the previous release runs again. Report error.message, error.details.log, and whether the database was restored; an external database may still contain changes from the failed upgrade. Leave the next attempt to the user once the cause is known.
- Exit 4: the application is down. Work through error.suggestions with the user, step by step.

rollback --dir <root> --json returns to the release the last upgrade came from, or to --to <release id or version>. When that upgrade migrated SQLite databases and the installer holds their backup, rolling back restores it and loses whatever was stored since; the confirmation notes say so, and the same relay-and-confirm step applies. External databases require a separate restore procedure: --backup-done does not make the installer restore them. If an irreversible migration made the schema incompatible with the previous runtime, returning to that runtime requires a compatible database backup too. --no-restore keeps the current databases; use it only when compatibility with the selected runtime is established.

Recover an interrupted operation

OPERATION_INTERRUPTED from upgrade, or pending in status, means an earlier run stopped while the application was down. Plain rollback recovers, through the same confirmation: it undoes an interrupted upgrade, restoring the databases, and finishes an interrupted rollback.

Change the origin or port

app-installer has no command for it. The origin, address and port live in app.env as APP_PUBLIC_ORIGIN, APP_SERVER_HOST and APP_SERVER_PORT. Before changing them, tell the user:

the application will be at the new origin plus its base path;
the application stops briefly while it restarts;
a new port has to be free, or the application restarts into a crash loop that only logs/app.err.log explains, and a reverse proxy in front has to forward to it;
APP_BASE_PATH cannot change this way: it is compiled into the client.

With their agreement, edit those three lines and nothing else, then run pm2 restart <name>, <name> being name in status. pm2 starts the application through launcher.mjs, which reads app.env on every start, so nothing has to be registered again. status afterwards reports the new endpoints and checks health at the address in app.env; for a new origin, have the user open endpoints.url.

Keep the data intact

The root holds the only copy of the application's configuration, secrets and data. Change everything in it through app-installer, apart from the three app.env lines above, and leave the choice of deleting anything to the user. Keep the contents of config.yml, which holds the secrets, out of the conversation.

When it activates

Automatically, when a task matches. Use when the user asks to deploy an application to production or a server, to upgrade or roll back such an installation, or when the directory holds an `installer.json` written by app-installer. Not for creating a project to develop, which is the `nocobase-create-app` Skill

About this skill

Automatic

DevOps & SRE pack

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v8 — 11/2/2026

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

nocobase / nocobase

skills/nocobase-app-installer/SKILL.md

View on GitHub ↗

Imported 11/2/2026Version v1Bundled 1 file

Illustrative Want to see what nocobase/nocobase-app-installer 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

Install and run NocoBase 3 with app-installer

Run it

Choose the route

Choose where the installer runs

Build the archive

Install

Every installation

Run it

When it succeeds

Several installations on one machine

Status

Upgrade and roll back

Recover an interrupted operation

Change the origin or port

Keep the data intact

When it activates

SkillSafety

Source

Related skills

More in the registry