Ready to supercharge your AI agents?

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

多智能体协同行业调研（OpenClaw 版） · leoyeai | DecimalAI

Researchby leoyeai

多智能体协同行业调研（OpenClaw 版）

Multi-agent collaborative industry research for OpenClaw. Dynamically assigns research roles, runs parallel research via sessions_spawn with codex/gemini/claude CLI augmentation, iterative quality review, merge & converge, outputs structured Markdown report. All parameters are configurable via environment variables or interactive setup. Trigger: /research, 行业调研, industry research, 调研报告

Never made a case worse

SkillScore

SkillScore82/100

Test results 82

Safety

3/3 safety checks passed

Code scan
Intent
Content

What it changes

23%82%of test cases passed +59 pts

Without skill5 of 22

With skill18 of 22

13 fixed5 passed either way4 still failing0 broke

Measured on gemini-3.6-flash8/8/2026v118 comparable

Track record

Agent runs: 0; All time

See one task, with and without the skill

One test taskIn the standard preset review phase, what is the maximum possible quality score, what total score threshold must be met to pass, and across how many evaluation dimensions is the score calculated?

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 leoyeai-industry-research
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("leoyeai-industry-research")
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 leoyeai-industry-research

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/leoyeai-industry-research/SKILL.md https://app.decimal.ai/s/leoyeai-industry-research/SKILL.md

Copy SKILL.md

Save it in your project, as

.claude/skills/leoyeai-industry-research/SKILL.md

This gets SKILL.md only. The skill's 2 other files are 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 leoyeai-industry-research

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("leoyeai-industry-research")
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/leoyeai-industry-research/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/leoyeai-industry-research/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 2 other files are 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 leoyeai-industry-research

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("leoyeai-industry-research")
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/leoyeai-industry-research/SKILL.md

Give it to your agent

from pathlib import Path
from agents import Agent, Runner

skill = Path(".agents/skills/leoyeai-industry-research/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 2 other files are 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 leoyeai-industry-research

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("leoyeai-industry-research")
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/leoyeai-industry-research/SKILL.md

Give it to your agent

from pathlib import Path
from openai import OpenAI

client = OpenAI()
skill = Path(".agents/skills/leoyeai-industry-research/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 2 other files are 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 leoyeai-industry-research

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("leoyeai-industry-research")
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/leoyeai-industry-research/SKILL.md

Give it to your agent

from pathlib import Path
import anthropic

client = anthropic.Anthropic()
skill = Path(".agents/skills/leoyeai-industry-research/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 2 other files are 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 leoyeai-industry-research

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("leoyeai-industry-research")
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/leoyeai-industry-research/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/leoyeai-industry-research/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 2 other files are 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 leoyeai-industry-research

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("leoyeai-industry-research")
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/leoyeai-industry-research/SKILL.md

Give it to your agent

from pathlib import Path
from langchain.agents import create_agent

skill = Path(".agents/skills/leoyeai-industry-research/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 2 other files are 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 leoyeai-industry-research

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("leoyeai-industry-research")
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/leoyeai-industry-research/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 2 other files are 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 leoyeai-industry-research

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("leoyeai-industry-research")
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/leoyeai-industry-research/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/leoyeai-industry-research/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 2 other files are 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 leoyeai-industry-research

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("leoyeai-industry-research")
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/leoyeai-industry-research/SKILL.md

Give it to your agent

from pathlib import Path
from pydantic_ai import Agent

skill = Path(".agents/skills/leoyeai-industry-research/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 2 other files are 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 leoyeai-industry-research

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("leoyeai-industry-research")
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/leoyeai-industry-research/SKILL.md

Give it to your agent

from pathlib import Path
from deepagents import create_deep_agent

skill = Path(".agents/skills/leoyeai-industry-research/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 2 other files are 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 leoyeai-industry-research

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("leoyeai-industry-research")
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/leoyeai-industry-research/SKILL.md

Give it to your agent

import asyncio
from pathlib import Path
from agent_framework import Agent

skill = Path(".agents/skills/leoyeai-industry-research/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 2 other files are 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 leoyeai-industry-research

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("leoyeai-industry-research")
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/leoyeai-industry-research/SKILL.md

Give it to your agent

from pathlib import Path
from strands import Agent

skill = Path(".agents/skills/leoyeai-industry-research/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 2 other files are 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 leoyeai-industry-research

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/leoyeai-industry-research/SKILL.md https://app.decimal.ai/s/leoyeai-industry-research/SKILL.md

Copy SKILL.md

Save it in your project, as

.agents/skills/leoyeai-industry-research/SKILL.md

This gets SKILL.md only. The skill's 2 other files are 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/leoyeai-industry-research/SKILL.md

Read the summary as JSON

curl -fsSL https://app.decimal.ai/s/leoyeai-industry-research.json

This gets SKILL.md only. The skill's 2 other files are not included.

Other ways to add it

Other setups docs

You don't need a DecimalAI account.

Version: v1, tested 8/8/2026
Source: LeoYeAI/openclaw-master-skills ↗
License: MIT

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

多智能体协同行业调研（OpenClaw 版）

协调多个 sub-agent 并行完成行业调研，通过迭代审核和合并收敛产出高质量报告。

Commands

/research <topic>                    # 启动行业调研（如：/research AI Agent 行业现状）
/research <topic> --depth=deep       # 深度调研模式
/research <topic> --depth=quick      # 快速概览模式

Prerequisites

OpenClaw — 主 agent 运行时（需启用 sessions_spawn 和 exec 工具）
Codex CLI (codex) — 可选，提供 OpenAI 模型补充视角
Gemini CLI (gemini) — 可选，提供 Google 模型补充视角
Claude CLI (claude) — 可选，提供 Anthropic 模型补充视角
CLI 工具未安装不影响核心流程，仅跳过补充视角

可配置参数

优先级：环境变量 > 启动时用户选择 > 默认值

参数	环境变量	默认值	说明
研究 agent 数量	`RESEARCH_AGENT_COUNT`	`auto`	设为数字则固定，`auto` 则动态分配
研究迭代上限	`RESEARCH_MAX_ROUNDS`	`3`	研究阶段最多迭代几轮
合并迭代上限	`MERGE_MAX_ROUNDS`	`2`	合并阶段最多迭代几轮
质量总分阈值	`QUALITY_THRESHOLD`	`35`	满分 50，总分需达到此值
单项最低分	`QUALITY_MIN_PER_DIM`	`6`	满分 10，每个维度不低于此值
交叉验证模式	`RESEARCH_CROSS_MODEL`	`cli`	`cli` / `native` / `both` / `none`，见下方说明
启用 CLI 工具	`RESEARCH_CLI_TOOLS`	`codex,gemini,claude`	可选值：`codex`、`gemini`、`claude`、`none` 或组合
CLI 超时	`RESEARCH_CLI_TIMEOUT`	`600`	秒
Native 模型列表	`RESEARCH_MODELS`	(空)	如 `anthropic/claude-sonnet-4-6,google/gemini-2.5-pro,openai-codex/gpt-5.4`
输出语言	`RESEARCH_LANG`	`zh`	`zh` 中文 / `en` 英文
报告深度	`RESEARCH_DEPTH`	`standard`	`quick` / `standard` / `deep`

交叉验证模式

不同 AI 模型有不同的知识覆盖、推理偏好和盲区。让多个模型分别分析同一问题，能有效提升调研质量。

模式	说明	适用场景
`cli`	研究 agent 在自己的分析完成后，必须调用 CLI 工具（codex/gemini/claude）对关键发现做交叉验证。CLI 使用订阅账号，无额外 API 费用。这是默认模式。	没有购买多个模型 API 的用户
`native`	不同研究 agent 通过 `sessions_spawn` 的 `model` 参数使用不同模型原生运行。每个 agent 拥有完整工具链。需要配置 `RESEARCH_MODELS`。	已购买多个模型 API key 的用户
`both`	结合两者：不同 agent 用不同模型 + agent 内部再用 CLI 交叉验证	追求最高质量的用户
`none`	所有 agent 使用默认模型，不做交叉验证	快速调研或预算有限

native 模式下的模型分配策略：

将 RESEARCH_MODELS 中的模型轮询分配给研究 agent
Reviewer 使用与多数研究 agent 不同的模型
Merger 使用与 Reviewer 不同的模型
示例：RESEARCH_MODELS=anthropic/claude-sonnet-4-6,google/gemini-2.5-pro,openai-codex/gpt-5.4

深度预设

参数	quick	standard	deep
Agent 数量	2-3	3-5	5-8
研究迭代上限	1	3	10
合并迭代上限	1	2	3
质量总分阈值	30	35	40
web_search 最少关键词数	3	5	10

工作流

阶段 0：配置加载与确认

收到调研主题后，先完成配置再开始调研：

检查历史调研残留（防污染）：用 exec 扫描工作目录下是否存在历史调研产出：
```
ls -d */round-1 */merged */output 2>/dev/null | sed 's|/.*||' | sort -u
```
- 如果发现历史调研目录：向用户列出这些目录及其创建时间，并询问：
  检测到以下历史调研目录，它们可能干扰本次调研（旧报告内容可能被误引用、旧 progress/task_plan 可能干扰状态判断）：
  - {dir1}/ ({date1})
  - {dir2}/ ({date2})
  是否清理？
  - A：归档到 _archive/ 后开始（推荐 — 移走但不删除）
  - B：直接删除后开始
  - C：保留，直接开始（不推荐）
- 同时检查工作目录根下的孤立状态文件（progress.md、task_plan.md、findings.md、tmp/、generated/），一并列入清理范围。
- 用户选 A 时：mkdir -p _archive && mv {dirs} _archive/，同时移走孤立状态文件。
- 用户选 B 时：rm -rf {dirs} 及孤立状态文件。
- 用户选 C 或无历史残留时：跳过，继续下一步。
- 重要：等待用户回复后再继续。
用 exec 读取环境变量：

echo "RESEARCH_DEPTH=${RESEARCH_DEPTH:-standard}"
echo "RESEARCH_AGENT_COUNT=${RESEARCH_AGENT_COUNT:-auto}"
echo "RESEARCH_MAX_ROUNDS=${RESEARCH_MAX_ROUNDS:-}"
echo "MERGE_MAX_ROUNDS=${MERGE_MAX_ROUNDS:-}"
echo "QUALITY_THRESHOLD=${QUALITY_THRESHOLD:-}"
echo "QUALITY_MIN_PER_DIM=${QUALITY_MIN_PER_DIM:-}"
echo "RESEARCH_CROSS_MODEL=${RESEARCH_CROSS_MODEL:-cli}"
echo "RESEARCH_CLI_TOOLS=${RESEARCH_CLI_TOOLS:-codex,gemini,claude}"
echo "RESEARCH_CLI_TIMEOUT=${RESEARCH_CLI_TIMEOUT:-600}"
echo "RESEARCH_MODELS=${RESEARCH_MODELS:-}"
echo "RESEARCH_LANG=${RESEARCH_LANG:-zh}"

如果设置了 RESEARCH_DEPTH，先应用对应预设，再用环境变量中的具体参数覆盖
检查 CLI 工具可用性：

command -v codex && codex --version || echo "CODEX_UNAVAILABLE"
command -v gemini && gemini --version || echo "GEMINI_UNAVAILABLE"
command -v claude && claude --version || echo "CLAUDE_UNAVAILABLE"

如果配置了某 CLI 但不可用，自动从 RESEARCH_CLI_TOOLS 中移除并提示用户。

检测当前 session 模型（模型继承）：子 agent 必须继承主 session 当前使用的模型，而非回退到 openclaw.json 默认值。你（主 agent）应当知道自己当前运行的模型 ID（例如 openai-codex/gpt-5.3-codex-spark、openai-codex/gpt-5.4、google/gemini-2.5-pro 等）。如果不确定，可用 exec 尝试获取：
```
curl -s http://localhost:18789/api/session/info 2>/dev/null | python3 -c "import json,sys; print(json.load(sys.stdin).get('model','UNKNOWN'))" || echo "UNKNOWN"
```
将模型 ID 记录为 inheritedModel，写入 config.md。 后续所有 sessions_spawn 调用必须传入 model: {inheritedModel}，确保子 agent 与主 session 使用同一模型。
向用户展示配置表并请求确认（直接输出，等待用户回复）：
- 选项 A：使用当前配置开始
- 选项 B：调整参数（逐项询问需修改的值）
- 选项 C：切换为深度调研模式
- 选项 D：切换为快速调研模式

重要：输出选项后停止，等待用户回复再继续执行。

创建项目目录，将最终配置写入 {topic-slug}/config.md：

mkdir -p "{topic-slug}/round-1" "{topic-slug}/merged" "{topic-slug}/output"

阶段 1 + 2：主题分析 → 立即 spawn 研究 agent

关键：阶段 1 和阶段 2 必须在同一个 turn 内连续完成，不得中间回复用户后停下。

用户确认配置后，在同一个 turn 中完成以下所有步骤：

分析主题：拆解调研领域，识别 3-8 个关键维度
动态生成角色，每个角色包含：
- 角色名称和代号（如 market-analyst、tech-trend）
- 调研范围和具体任务描述
- 关键问题清单（3-5 个必答问题）
- 推荐 CLI 工具（Codex/Gemini/Claude/仅 web_search）
用 write 写入 brief.md
不要停下，不要等待用户确认角色方案，立即执行下一步：
立即用 sessions_spawn 逐个启动 N 个研究 agent（每个用不同 label）。这是必须调用的工具，不可跳过：

cli 模式（默认）：所有 agent 继承主 session 模型，prompt 中包含 CLI 交叉验证指令

sessions_spawn({ task: "...(含 CLI 交叉验证指令)", label: "researcher-market-analyst", model: "{inheritedModel}", mode: "run", runTimeoutSeconds: 900 })
sessions_spawn({ task: "...(含 CLI 交叉验证指令)", label: "researcher-tech-trend", model: "{inheritedModel}", mode: "run", runTimeoutSeconds: 900 })

native 模式：每个 agent 指定不同 model（从 RESEARCH_MODELS 轮询分配）

sessions_spawn({ task: "...", label: "researcher-market-analyst", model: "anthropic/claude-sonnet-4-6", mode: "run", runTimeoutSeconds: 900 })
sessions_spawn({ task: "...", label: "researcher-tech-trend", model: "google/gemini-2.5-pro", mode: "run", runTimeoutSeconds: 900 })
sessions_spawn({ task: "...", label: "researcher-app-scenario", model: "openai-codex/gpt-5.4", mode: "run", runTimeoutSeconds: 900 })

both 模式：指定不同 model + prompt 中包含 CLI 交叉验证指令

用 write 将所有返回的 {runId, childSessionKey, label} 记录到 {topic-slug}/round-{N}/tracking.md
回复用户通知："已启动 N 个研究 agent，正在并行调研，完成后我会汇总。"
等待 announce 消息：每个 sub-agent 完成后会自动发送 announce 回主 session。收到 announce 时：
- 用 read 读取该 agent 的产出文件 {topic-slug}/round-{N}/agent-{role-slug}.md
- 将产出的"核心发现"和"交叉验证小结"章节内容发送给用户，格式如：
  ✅ Agent {label} 已完成 ({completed}/{total}) 核心发现：
  1. xxx
  2. xxx 交叉验证： {一致/矛盾要点}
- 这样用户可以实时检查每个 agent 的产出质量
兜底机制：如果等待超过合理时间，用 sessions_list 检查未完成 agent 状态
收到所有 announce 后（包括失败的），立即执行以下操作，不得停下：
- 用 exec 检查 {topic-slug}/round-{N}/ 下有哪些 agent-*.md 文件
- 统计成功/失败数量，通知用户
- 如果 ≥50% agent 有产出 → 立即 spawn reviewer（进入阶段 3）
- 不要说"我会接着收口"然后停下——这会中断整个流程

阶段 3：审核与迭代

用 sessions_spawn 启动 reviewer agent：

sessions_spawn：
  task: {Reviewer Prompt}
  label: "reviewer-round-{N}"
  model: "{inheritedModel}"
  mode: "run"
  runTimeoutSeconds: 600

等待 reviewer 的 announce 消息
用 read 读取 {topic-slug}/round-{N}/review.md，将评审结果发送给用户：

📋 第 {N} 轮评审结果：{通过/不通过} — 总分 {X}/50

维度得分
（各维度评分表）
需改进： {关键修改指令摘要}
根据评分决策：
- 通过（总分 ≥ QUALITY_THRESHOLD 且单项 ≥ QUALITY_MIN_PER_DIM）→ 进入阶段 4
- 不通过且轮次 < RESEARCH_MAX_ROUNDS → 回到阶段 2（spawn 新 agent，prompt 中包含上轮产出 + 反馈）
- 不通过且已达上限 → 通知用户并等待决策

维度	得分
（各维度评分表）
需改进： {关键修改指令摘要}

阶段 4：合并收敛

用 sessions_spawn 启动 merger agent：

sessions_spawn：
  task: {Merger Prompt}
  label: "merger-{M}"
  model: "{inheritedModel}"
  mode: "run"
  runTimeoutSeconds: 900

等待 merger 的 announce，用 read 读取合并稿，将执行摘要发送给用户：

📄 合并稿已完成，执行摘要： {执行摘要内容}
用 sessions_spawn 启动 reviewer 审核合并稿，写入 {topic-slug}/merged/review-{M}.md
将合并稿审核结果发送给用户（同阶段 3 格式）
判断逻辑同阶段 3，上限为 MERGE_MAX_ROUNDS

阶段 5：最终输出

用 read 读取通过审核的合并稿
最终格式化和润色
用 write 写入 {topic-slug}/output/report.md
向用户展示报告摘要和文件路径

Agent Prompt 模板

研究 Agent Prompt

spawn 每个研究 agent 时，使用以下 prompt（替换 {...} 变量）：

你是行业调研团队的 {角色名}（{角色代号}）。

## 你的任务
{来自 brief.md 的角色定义和任务描述}

## 必答问题
{来自 brief.md 的关键问题清单，每个问题单独一行}

## 工作流程
1. 使用 web_search 搜索相关信息（至少搜索 {config.minSearchKeywords} 个不同关键词，覆盖中英文）
2. 整理初步研究结果，识别核心发现
3. 调用 {推荐的 CLI 工具} 获取补充视角（遵循下方 CLI 调用协议）
4. 按照输出模板格式，用 write 将研究报告写入文件：{目标文件路径}
5. 完成后返回结果

{如果是迭代轮次 > 1，包含以下段落：}
## 上轮产出和反馈
请先用 read 阅读以下文件：
- 你的上轮产出：{上轮文件路径}
- Reviewer 反馈：{review 文件路径}

重点关注 reviewer 对你（{角色名}）的具体修改指令，逐条落实。保留上轮中已获好评的内容，只改进被指出的问题。
{/如果}

## 输出模板（必须严格遵循此格式）

# {调研维度名称}

> 调研员：{角色名} | 轮次：{N} | 日期：{今日日期}

## 核心发现
1. {发现1 — 一句话概括}
2. {发现2}
3. {发现3}
（3-5 条）

## 详细分析

### {子主题 1}
{分析内容，包含数据、趋势、案例}

### {子主题 2}
{分析内容}

## 数据与信源
| 数据点 | 数值 | 来源 | 可信度 |
|--------|------|------|--------|
| {数据名} | {具体数值} | {来源名称和链接} | {高/中/低} |

## 交叉验证（CLI 模式下必填，native/none 模式删除此节）
> 以下为通过不同 AI 模型对核心发现的交叉验证结果

### {CLI 工具名} 验证结果
{CLI 返回的分析 — 重点标注与你的发现一致或矛盾的部分}

### 交叉验证小结
| 核心发现 | 本 agent 结论 | CLI 验证结论 | 是否一致 |
|----------|--------------|-------------|---------|
| {发现1} | {你的结论} | {CLI 结论} | 一致/矛盾/补充 |

## 风险与不确定性
- {不确定因素1}
- {不确定因素2}

## 待深入研究
- {可进一步挖掘的方向}

## CLI 交叉验证（当 RESEARCH_CROSS_MODEL 为 cli 或 both 时，此步骤必须执行，不可跳过）

完成 web_search 和初步分析后，**你必须用 exec 调用 CLI 工具对核心发现做交叉验证**。详细步骤见 references/cross-validation.md。

简要流程：
1. `exec`: 将核心发现写入 `/tmp/research-cross-validate.txt`
2. `exec`: 调用推荐 CLI（如 `cat /tmp/research-cross-validate.txt | gemini -p "审核提示..." 2>&1`）
3. 失败则按降级链尝试下一个 CLI，全部失败标注 "[交叉验证：CLI 不可用]"
4. 将 CLI 返回整合到"交叉验证"章节
5. 清理临时文件

## 规则
- 数据必须标注来源和可信度评估
- 不要编造数据，找不到就明确标注"未找到可靠数据"
- 对不确定的判断使用"可能""推测""初步判断"等措辞
- 保持客观中立，呈现多元视角
- 优先引用一手数据和权威信源

Reviewer Agent Prompt

你是行业调研团队的质量审核员。你的职责是严格、公正地评估调研产出的质量。

## 你的任务
审核第 {N} 轮所有调研 agent 的产出，按评分标准评审，给出结构化反馈。

## 评审标准（5 维度，各 10 分，满分 50）

| 维度 | 评分标准 |
|------|----------|
| 事实准确性 | 声明是否有据可查、数据是否标注来源、是否存在明显错误或自相矛盾 |
| 信源覆盖度 | 是否涵盖多元视角（中英文、不同机构）、是否有一手/权威信源、是否存在明显遗漏领域 |
| 分析深度 | 是否有原因分析而非仅罗列事实、是否有独到洞察、逻辑推理是否严密 |
| 内部一致性 | 各 agent 之间数据是否矛盾、同一 agent 内部论述是否自洽、结论是否与论据匹配 |
| 完整性 | 相对于 brief.md 中定义的调研范围和必答问题，覆盖率如何、有无关键遗漏 |

## 通过标准
总分 ≥ {config.qualityThreshold}/50，且每个单项 ≥ {config.qualityMinPerDim}/10

## 工作流程
1. 用 read 阅读调研简报：{brief.md 路径}
2. 用 read 逐个阅读所有 agent 产出：{各文件路径列表}
3. 按 5 个维度逐一打分
4. 交叉验证各 agent 之间的数据一致性
5. 对每个 agent 写出具体、可执行的修改指令
6. 做出通过/不通过决定
7. 用 write 按以下模板写入文件：{review.md 路径}
8. 完成后返回结论（通过/不通过 + 总分）

## 输出模板（必须严格遵循）

# 第 {N} 轮评审报告

## 评审结论：{通过 / 不通过}
## 总分：{X}/50

## 各维度评分
| 维度 | 得分 | 说明 |
|------|------|------|
| 事实准确性 | {X}/10 | {一句话说明扣分/加分原因} |
| 信源覆盖度 | {X}/10 | {一句话说明} |
| 分析深度 | {X}/10 | {一句话说明} |
| 内部一致性 | {X}/10 | {一句话说明} |
| 完整性 | {X}/10 | {一句话说明} |

## 亮点
- {值得保留的优质内容}

## 逐 Agent 修改指令

### Agent: {角色名}
1. {具体修改指令 — 明确说明需要补充/修改什么、期望的产出是什么}
2. {具体修改指令}

### Agent: {角色名}
1. {具体修改指令}

## 整体建议
{对整体调研方向或方法的建议，如有}

## 规则
- 评分必须基于实际内容质量，不可随意给高分放水
- 修改指令必须具体可执行（不可只说"需要改进"，要说明改进什么、怎么改）
- 交叉验证各 agent 之间的数据，发现矛盾必须指出
- 关注信源的权威性、时效性和多元性
- 如果某 agent 产出已经很好，明确说明"保持现状"而非无病呻吟

Merger Agent Prompt

你是行业调研团队的报告整合专家。你的职责是将多个调研 agent 的产出整合为一份完整、连贯、有洞察的行业调研报告。

## 你的任务
将以下 agent 的调研产出合并为一份统一报告。

## 工作流程
1. 用 read 阅读调研简报：{brief.md 路径}
2. 用 read 阅读调研配置：{config.md 路径}
3. 用 read 逐个阅读所有 agent 最终产出：{各文件路径列表}
4. 提取各维度的核心发现
5. 识别并处理各 agent 之间的矛盾或重叠
6. 构建完整叙事逻辑，形成有洞察的报告
7. 用 write 按以下结构输出到：{merged/draft-N.md 路径}
8. 完成后返回结果

{如果是修订轮，包含以下段落：}
## Reviewer 反馈
请先用 read 阅读 reviewer 的审核意见：{review 文件路径}
按反馈意见逐条修订报告。保留已获好评的内容。
{/如果}

## 报告结构

# {调研主题} — 行业调研报告

> 日期：{今日日期} | 调研深度：{config.depth} | 迭代轮次：{实际轮次}

## 执行摘要
{500 字以内，概括核心发现、关键数据和主要结论}

## 1. 行业概述
{行业定义、边界、发展历程简述}

## 2. {维度 1 标题}
{整合该维度所有 agent 的研究成果}

## 3. {维度 2 标题}
{整合内容}

...（按实际维度数量扩展）

## N. 关键发现与洞察
{跨维度的核心洞察，不是简单罗列而是有分析的判断}

## N+1. 风险与挑战
{整合所有 agent 提到的风险，去重并分类}

## N+2. 展望与建议
{基于分析得出的行动建议}

## 附录
### 数据来源汇总
{所有信源的汇总表}

### 调研方法说明
{本次调研使用的方法、工具、agent 角色说明}

## 规则
- 去重合并时保留最准确、最有信源支持的版本
- 对矛盾数据必须标注说明（"A 数据显示 X，而 B 数据显示 Y"），不可静默丢弃
- 保持报告的叙事连贯性，避免简单拼接各 agent 的内容
- 所有数据保留原始信源标注
- 执行摘要必须能独立阅读，包含最重要的 3-5 个发现

主 Agent 编排逻辑

收到 /research <topic> 后：

阶段 0（可与用户交互）：exec 读取环境变量 → 应用预设 → 检查 CLI → 展示配置等待用户确认→ write config.md
阶段 1+2（必须在一个 turn 内连续完成，不可中途停下）：分析主题 → 生成角色 → write brief.md → 立即调用 sessions_spawn 启动所有研究 agent → write tracking.md → 通知用户已启动
阶段 3：收到所有 announce 后 → sessions_spawn 启动 reviewer → 等待 announce → read review.md → 通过/不通过决策
阶段 4：sessions_spawn 启动 merger → announce 后 spawn reviewer 审核
阶段 5：read 合并稿 → 润色 → write output/report.md → 展示摘要

核心规则：

模型继承：所有模式下，sessions_spawn 都必须带 model 参数。cli/none 模式传入 {inheritedModel}（从主 session 继承）；native/both 模式从 RESEARCH_MODELS 轮询分配。绝不可省略 model 参数让子 agent 回退到 openclaw.json 默认值。
只有阶段 0 允许等待用户回复。阶段 1 到阶段 5 应连续执行。
cli/both 模式下：研究 agent 的 prompt 必须包含 CLI 交叉验证协议，且协议中明确标注"必须执行，不可跳过"。
native/both 模式下：sessions_spawn 必须带 model 参数，从 RESEARCH_MODELS 轮询分配。Reviewer 和 Merger 使用与多数研究 agent 不同的模型。

异步 Agent 异常处理与自动推进规则

核心原则：收到所有 announce 后（无论成功还是失败），必须立即推进到下一阶段，不得停下。

自动推进规则

收到所有 N 个 agent 的 announce 后，检查产出文件：

≥50% 的 agent 有产出 → 立即 spawn reviewer，在 reviewer prompt 中标注缺失的维度
<50% 有产出 → 通知用户，询问是否重试失败的 agent 或降级继续
全部失败 → 通知用户并建议排查

关键：不要因为某个 agent 失败就停下来等待或"接着收口"。失败是正常的，用现有产出继续推进流程。

失败 agent 的处理

检查失败 agent 的输出文件是否存在（可能 abort 前已部分写入）
有部分产出 → 在 reviewer prompt 中标注 "[不完整，仅供参考]"
无产出 → 在 reviewer prompt 中标注 "维度 {X} 缺失（agent 执行失败），请在评审中指出此缺失对整体质量的影响"
在 merged 报告中对缺失维度标注说明，不要假装它存在

文件结构

每次调研在当前工作目录下创建：

{topic-slug}/
├── config.md                    # 本次调研的实际配置
├── brief.md                     # 调研简报（主题、范围、角色分配）
├── round-1/
│   ├── tracking.md              # sub-agent 跟踪信息
│   ├── agent-{role-slug}.md     # 各 agent 调研产出
│   └── review.md               # Reviewer 评审结果
├── round-2/                     # 如需迭代
│   ├── tracking.md
│   ├── agent-{role-slug}.md
│   └── review.md
├── merged/
│   ├── draft-1.md               # 合并稿
│   ├── review-1.md              # 合并稿审核
│   └── ...
└── output/
    └── report.md               # 最终报告

About this skill

Automatic

Research pack

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/2/2026

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

LeoYeAI / openclaw-master-skills

skills/multi-agent-industry-research/SKILL.md

View on GitHub ↗

License MITImported 8/2/2026Version v1Bundled 2 files

Illustrative Want to see what leoyeai/industry-research 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

多智能体协同行业调研（OpenClaw 版）

Commands

Prerequisites

可配置参数

交叉验证模式

深度预设

工作流

阶段 0：配置加载与确认

阶段 1 + 2：主题分析 → 立即 spawn 研究 agent

阶段 3：审核与迭代

阶段 4：合并收敛

阶段 5：最终输出

Agent Prompt 模板

研究 Agent Prompt

Reviewer Agent Prompt

Merger Agent Prompt

主 Agent 编排逻辑

异步 Agent 异常处理与自动推进规则

自动推进规则

失败 agent 的处理

文件结构

SkillSafety

Source

Related skills

More in the registry