Ready to supercharge your AI agents?

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

Get Started Free →

Meta Facebook Inbox · LeoYeAI | DecimalAI

Sales & Supportby LeoYeAI

Meta Facebook Inbox

Check Facebook page inbox messages via Meta Business Suite browser automation.

SkillScore

SkillScore82/100

Test results 64 · Live runs 100

Safety

3/3 safety checks passed

Code scan
Intent
Content

What it changes

5%64%of test cases passed +59 pts

Without skill1 of 22

With skill14 of 22

14 fixed0 passed either way7 still failing1 broke

Measured on gemini-3.6-flash7/28/2026v119 comparable

Track record

Agent runs: 14; All time

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 meta-fb-inbox
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("meta-fb-inbox")
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 meta-fb-inbox

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/meta-fb-inbox/SKILL.md https://app.decimal.ai/s/meta-fb-inbox/SKILL.md

Copy SKILL.md

Save it in your project, as

.claude/skills/meta-fb-inbox/SKILL.md

This gets SKILL.md only. The skill's 3 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 meta-fb-inbox

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("meta-fb-inbox")
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/meta-fb-inbox/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/meta-fb-inbox/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 3 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 meta-fb-inbox

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("meta-fb-inbox")
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/meta-fb-inbox/SKILL.md

Give it to your agent

from pathlib import Path
from agents import Agent, Runner

skill = Path(".agents/skills/meta-fb-inbox/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 3 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 meta-fb-inbox

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("meta-fb-inbox")
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/meta-fb-inbox/SKILL.md

Give it to your agent

from pathlib import Path
from openai import OpenAI

client = OpenAI()
skill = Path(".agents/skills/meta-fb-inbox/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 3 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 meta-fb-inbox

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("meta-fb-inbox")
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/meta-fb-inbox/SKILL.md

Give it to your agent

from pathlib import Path
import anthropic

client = anthropic.Anthropic()
skill = Path(".agents/skills/meta-fb-inbox/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 3 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 meta-fb-inbox

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("meta-fb-inbox")
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/meta-fb-inbox/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/meta-fb-inbox/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 3 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 meta-fb-inbox

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("meta-fb-inbox")
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/meta-fb-inbox/SKILL.md

Give it to your agent

from pathlib import Path
from langchain.agents import create_agent

skill = Path(".agents/skills/meta-fb-inbox/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 3 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 meta-fb-inbox

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("meta-fb-inbox")
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/meta-fb-inbox/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 3 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 meta-fb-inbox

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("meta-fb-inbox")
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/meta-fb-inbox/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/meta-fb-inbox/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 3 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 meta-fb-inbox

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("meta-fb-inbox")
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/meta-fb-inbox/SKILL.md

Give it to your agent

from pathlib import Path
from pydantic_ai import Agent

skill = Path(".agents/skills/meta-fb-inbox/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 3 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 meta-fb-inbox

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("meta-fb-inbox")
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/meta-fb-inbox/SKILL.md

Give it to your agent

from pathlib import Path
from deepagents import create_deep_agent

skill = Path(".agents/skills/meta-fb-inbox/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 3 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 meta-fb-inbox

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("meta-fb-inbox")
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/meta-fb-inbox/SKILL.md

Give it to your agent

import asyncio
from pathlib import Path
from agent_framework import Agent

skill = Path(".agents/skills/meta-fb-inbox/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 3 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 meta-fb-inbox

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("meta-fb-inbox")
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/meta-fb-inbox/SKILL.md

Give it to your agent

from pathlib import Path
from strands import Agent

skill = Path(".agents/skills/meta-fb-inbox/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 3 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 meta-fb-inbox

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/meta-fb-inbox/SKILL.md https://app.decimal.ai/s/meta-fb-inbox/SKILL.md

Copy SKILL.md

Save it in your project, as

.agents/skills/meta-fb-inbox/SKILL.md

This gets SKILL.md only. The skill's 3 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/meta-fb-inbox/SKILL.md

Read the summary as JSON

curl -fsSL https://app.decimal.ai/s/meta-fb-inbox.json

This gets SKILL.md only. The skill's 3 other files are not included.

Other ways to add it

Other setups docs

You don't need a DecimalAI account.

Version: v1, tested 7/28/2026
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

Meta Facebook Inbox

⚠️ CRITICAL: Always use profile:"openclaw" (isolated browser) for all browser actions. Never use Chrome relay. Keep the same targetId across operations to avoid losing the tab.

⚠️ CONTEXT Management:

Snapshot results are large. Do NOT repeat them in thinking.
Extract needed info and process immediately.
Preferred approach: Use snapshot refs:"aria" compact:true to find element refs, then act + click/type with ref.

Configuration

Before first use, check if config.json exists:

read file_path:"skills/meta-fb-inbox/config.json"

⚠️ PATH RULE: Always use workspace-relative paths starting from skills/meta-fb-inbox/. Never use ../ or absolute paths.

If missing or empty, help the user run the setup wizard:

cd skills/meta-fb-inbox
node scripts/setup.js

Config Format

config.json contains a pages array. Each page has an alias (display name) and url (Meta Business Suite inbox URL):

{
  "pages": [
    {
      "alias": "fb fanpage",
      "url": "https://business.facebook.com/latest/inbox/all/?&asset_id=123456789012345"
    },
    {
      "alias": "fb fanpage 2",
      "url": "https://business.facebook.com/latest/inbox/all/?&asset_id=987654321098765"
    }
  ]
}

Resolving a Page

When the user asks to check messages:

If they specify an alias (e.g. "fb fanpage 2"), look it up in config.json → pages.
If they don't specify, use the first page in the array.
If only one page exists, use it directly.
If multiple pages exist and no alias given, list available aliases and ask which one.

Enter Facebook Inbox

This is the first step for all operations below.

Check and start browser service:
```
browser action:"status"
```
If "running": false, start the browser service:
```
browser action:"start" profile:"openclaw"
```
Wait 2-3 seconds for the service to initialize.
Read config to get the target page URL:
```
read file_path:"skills/meta-fb-inbox/config.json"
```
Resolve the page alias to a URL (see "Resolving a Page" above).

Open browser:

browser action:"open" profile:"openclaw" targetUrl:"<pageUrl>"

Capture targetId and url from response.

Wait for page load:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"wait","timeMs":3000}

Check login state:
```
browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"evaluate","fn":"function(){return window.location.href;}"}
```
- If URL contains business.facebook.com and shows the inbox → logged in, proceed.
- If redirected to a login page → notify the user they need to log in manually in the OpenClaw browser, then retry.
Proceed with the user's request.

Clean Up Browser Tabs (after all operations):

a. Open a blank tab:

browser action:"open" profile:"openclaw" targetUrl:"about:blank"

b. Get all tabs:

browser action:"tabs" profile:"openclaw"

c. Close all tabs except the newest one (about:blank):

browser action:"close" profile:"openclaw" targetId:"<your_targetId>"

Quick Check (for cron / automated checks)

Complete flow: open page → screenshot → report.

Read config: read file_path:"skills/meta-fb-inbox/config.json" → resolve page URL
Start browser if needed: browser action:"start" profile:"openclaw"
Open page: browser action:"open" profile:"openclaw" targetUrl:<pageUrl> → get targetId
Wait: browser action:"act" profile:"openclaw" targetId:<targetId> request:{"kind":"wait","timeMs":4000}
Check URL to verify login (see step 5 above)
Take snapshot: browser action:"snapshot" profile:"openclaw" targetId:<targetId> refs:"aria" compact:true
Look for conversation items in the inbox list. Each conversation typically shows:
- Customer name
- Last message preview
- Timestamp
- Unread indicator (bold text or dot)
Report results in format: <name> (<time>) <preview> [未讀/已讀]
Clean up tabs.

If the snapshot is hard to parse (Meta's DOM is complex), fall back to:

browser action:"screenshot" profile:"openclaw" targetId:<targetId>

Then describe what you see in the screenshot.

Check Inbox Messages

Follow "Enter Facebook Inbox" steps to get to the inbox page.
Take a snapshot or screenshot to see the conversation list.
Report conversations with name, time, preview, and read/unread status.

Get URLs for Conversations (取得對話網址 - Simple!)

Why get URLs? Having the direct conversation URL lets you jump straight to a chat later without searching for it again. Saves time when replying.

When to get URLs: After listing chats, get URLs for unread conversations or any conversation you might need to access again.

📝 Simple 3-Step Method:

Click into the conversation using the customer name:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"evaluate","fn":"function() { const containers = document.querySelectorAll('div[style*=\"position: absolute\"]'); for (let container of containers) { const nameEl = container.querySelector('div.x1vvvo52.x1fvot60.xxio538'); if (nameEl && nameEl.textContent.trim() === '<customer_name>') { nameEl.click(); return {clicked: true}; } } return {error: 'not found'}; }"}

Replace <customer_name> with the actual customer name from your conversation list.

Wait 2 seconds for page to load and URL to update:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"wait","timeMs":2000}

Get the current URL - it's automatically updated!

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"evaluate","fn":"function(){return window.location.href;}"}

Store the URL alongside the chat info.

✅ Example URL Format:

https://business.facebook.com/latest/inbox/all/?&asset_id=123456789012345&selected_item_id=1234567890&thread_type=FB_MESSAGE

Or with optional mailbox_id:

https://business.facebook.com/latest/inbox/all/?&asset_id=123456789012345&mailbox_id=123456789012345&selected_item_id=9876543210&thread_type=FB_MESSAGE

🔍 Understanding the URL:

asset_id = Your Facebook page ID (stays the same)
selected_item_id = Unique conversation ID (different for each customer)
thread_type=FB_MESSAGE = Messenger conversation type
mailbox_id = Optional, may appear for some conversations

💾 How to Use Later:

Next time you need to access this conversation, skip all the searching:

browser action:"navigate" profile:"openclaw" targetId:"<targetId>" targetUrl:"<saved_conversation_url>"

Wait 2 seconds, and you're in the conversation!

💡 Tip: Get URLs for unread messages during your check routine, store them in a file or memory, and reuse them for instant access.

🔄 Getting Multiple URLs:

If you need URLs for multiple conversations:

Get URL for first conversation (steps 1-3 above)

Navigate back to inbox list:

browser action:"navigate" profile:"openclaw" targetId:"<targetId>" targetUrl:"<pageUrl>"

Wait 2 seconds.

Repeat for next conversation.

Alternative method (when already in conversation): If you're already viewing a conversation and just need its URL, skip step 1 and directly run step 3 - the URL is already there!

View Specific Conversation

⚡ Fast Path: If you have the conversation URL from "Get URLs for Conversations":

browser action:"navigate" profile:"openclaw" targetId:"<targetId>" targetUrl:"<conversation_url>"

Wait 2 seconds, then skip to step 3.

Standard Path:

From the inbox list, find and click the conversation (use snapshot + click by ref).
Wait for messages to load (2-3 seconds).
Take a screenshot to see the message thread.
Report the messages with sender, time, and content.

Extract Messages with Images

To programmatically read messages:

Navigate to the conversation (if not already there).

Inject the read-messages script:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"evaluate","fn":"<contents_of_scripts/read-messages.js>"}

Read the script file:

read file_path:"skills/meta-fb-inbox/scripts/read-messages.js"

And pass its entire contents as the fn parameter (wrapped in parentheses).

Parse the response: The script returns an array of message objects:

[
  {"text": "你好", "isCustomer": true, "hasImage": false, "imageUrl": null},
  {"text": "[Image]", "isCustomer": true, "hasImage": true, "imageUrl": "https://scontent-..."}
]

Download images (if any):

⚠️ Default download location: ~/Downloads (do NOT clutter the workspace).

For each message with hasImage: true:
```
cd ~/Downloads
curl -O "<imageUrl>"
```
The downloaded file will use the original filename from the URL.

If you want a custom filename:
```
cd ~/Downloads
curl -o "fb-message-$(date +%Y%m%d-%H%M%S).jpg" "<imageUrl>"
```

Reply to a Message

⚡ Fast Path (if you have conversation URL):

If you already obtained the conversation URL from "Get URLs for Conversations" section, skip step 1 and go directly:

browser action:"navigate" profile:"openclaw" targetId:"<targetId>" targetUrl:"<conversation_url>"

Wait 2 seconds, then proceed to step 2 (Take snapshot to find input box).

Standard Path (when you don't have the URL):

Open the conversation:

Click the customer name in the conversation list:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"evaluate","fn":"function() { const containers = document.querySelectorAll('div[style*=\"position: absolute\"]'); for (let container of containers) { const nameEl = container.querySelector('div.x1vvvo52.x1fvot60.xxio538'); if (nameEl && nameEl.textContent.trim() === '<customer_name>') { nameEl.click(); return {clicked: true}; } } return {error: 'not found'}; }"}

Replace <customer_name> with the actual customer name.

Wait 2 seconds for the conversation to load.

Take a snapshot to find the input box:

browser action:"snapshot" profile:"openclaw" targetId:"<targetId>" refs:"aria" compact:true

Look for textbox with [active] attribute.

Type the reply message:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"type","ref":"<input_ref>","text":"<your_message>"}

After typing, wait 1 second for the send button to appear.

Take another snapshot to find the send button:
```
browser action:"snapshot" profile:"openclaw" targetId:"<targetId>" refs:"aria" compact:true
```
Look for a new button near the textbox. The send button appears after you type text (it replaces the "like" button).

Click the send button:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"click","ref":"<send_button_ref>"}

Verify the message was sent:

Wait 2 seconds, then take a screenshot or check the conversation list to confirm the message appears.

Manage Labels

⚡ Fast Path: If you have the conversation URL, navigate directly:

browser action:"navigate" profile:"openclaw" targetId:"<targetId>" targetUrl:"<conversation_url>"

Wait 2 seconds, then proceed to the label operations.

Remove a Label

Open the conversation (if not already open, and you don't have URL).
Take a snapshot to locate the label's remove button:
```
browser action:"snapshot" profile:"openclaw" targetId:"<targetId>" refs:"aria" compact:true
```
Look for the Labels section in the right sidebar (appears before Notes section):
- A heading with level=3 (the Labels heading)
- Each label appears as a button element
- Inside each label button, there's a nested button with text "clearLabel"
- The clearLabel button is the "×" (remove) icon

Click the clearLabel button:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"click","ref":"<clearLabel_ref>"}

Example from snapshot:

- heading [level=3] [ref=eAAA]        ← Labels heading
- button [ref=e949]:                  ← Label button (shows label name)
    - button "clearLabel" [ref=e952]  ← Remove button (THIS ONE!)

Click the clearLabel button (ref=e952) to remove the label.

Wait for the change to apply:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"wait","timeMs":1500}

Verify the label was removed (optional): Take a screenshot or snapshot to confirm the label is gone.

Add a Label

Open the conversation.
Take a snapshot to find the label input combobox:
```
- combobox "label" [ref=eXXX]
```

Type the label name:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"type","ref":"<combobox_ref>","text":"<label_name>","submit":true}

Wait and verify.

Manage Notes (備註)

⚡ Fast Path: If you have the conversation URL, navigate directly:

browser action:"navigate" profile:"openclaw" targetId:"<targetId>" targetUrl:"<conversation_url>"

Wait 2 seconds, then proceed to the note operations.

Quick Decision Guide (看這裡！)

Question: Do you need to ADD or EDIT a note?

Take a snapshot of the conversation sidebar.
Look for the Notes section (usually appears after Labels section).
Check what you see:
- ONLY see a "Add note" type button/link (single button, no existing note text) → Contact has NO note → Use "Add a New Note" section below.
- See an "Edit" link + existing note text (usually with timestamp) → Contact has existing note → Use "Edit an Existing Note" section below.

Simple rule:

No note = Add new (only one button visible)
Has note = Edit existing (edit link + note text visible)

How to identify in snapshot:

No note: Look for pattern like button [ref=eXXX] or link [ref=eXXX] with NO paragraph text nearby
Has note: Look for pattern like link [ref=eXXX] (edit) + link [ref=eYYY] (delete) + paragraph [ref=eZZZ]: "note text"

Note Structure

The Notes section is located below the Labels section in the right sidebar. There are two states:

No note exists: You will see a single button or link (typically the "add note" action).
Note exists: You will see:
- The note text (in a paragraph element)
- An Edit link (usually appears before the note text with timestamp)
- A Delete link (next to edit link)
- Sometimes an "add note" link (for additional notes, but we prefer single-note approach)

Best Practice: Keep only one note per contact. Use "Edit" to update existing notes rather than adding multiple notes.

In snapshot, look for these patterns:

No note:

- heading [level=3] [ref=eXXX]     ← Notes heading
- button [ref=eYYY]                 ← Single "add note" button

Has note:

- heading [level=3] [ref=eXXX]     ← Notes heading
- link [ref=eYYY]                   ← "add note" link
  - text: "X minutes ago ·"        ← Timestamp
  - link [ref=eZZZ]                 ← Edit link (THIS ONE for editing!)
  - text: "·"
  - link [ref=eAAA]                 ← Delete link
- paragraph [ref=eBBB]: "note text" ← Existing note content

Edit an Existing Note (編輯註記 - Simple Steps for AI)

When to use: Contact already has a note. You will see an Edit link next to existing note text with timestamp.

Step-by-step:

Open the conversation (click customer name in chat list). Wait 2 seconds for it to load.

Take snapshot to find the Edit button:

browser action:"snapshot" profile:"openclaw" targetId:"<targetId>" refs:"aria" compact:true

Look for the pattern (Notes section):

- heading [level=3] [ref=eAAA]           ← Notes heading
- link [ref=eBBB]                        ← "add note" link
  - text: "4 hours ago ·"                ← Timestamp
  - link [ref=eXXX]                      ← Edit link (THIS ONE!)
  - text: "·"
  - link [ref=eYYY]                      ← Delete link
- paragraph [ref=eZZZ]: "Customer note" ← Current note

Key: Find the link that appears AFTER the timestamp text and BEFORE the "·" separator. That's your Edit button.

Click Edit button:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"click","ref":"<edit_button_ref>"}

Wait for edit form:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"wait","timeMs":1500}

Take snapshot to find the textbox:

browser action:"snapshot" profile:"openclaw" targetId:"<targetId>" refs:"aria" compact:true

Look for:

- textbox [ref=eXXX]: "Customer note"  ← Current text (contains existing note)
- button [ref=eYYY]                     ← Cancel button
- button [ref=eZZZ]                     ← Delete button
- button [disabled] [ref=eAAA]          ← Save button (disabled until you change text)

Key: The textbox will have the current note text. The Save button will show [disabled] attribute.

Click textbox to focus:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"click","ref":"<textbox_ref>"}

Select all existing text (Cmd+A):

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"press","key":"Meta+a"}

Type new text (replaces selected text):

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"type","ref":"<textbox_ref>","text":"Customer ID: 12345 ✅"}

Wait for Save button to enable:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"wait","timeMs":500}

Take snapshot to find enabled Save button:

browser action:"snapshot" profile:"openclaw" targetId:"<targetId>" refs:"aria" compact:true

Look for:

- button [ref=eXXX] [cursor=pointer]:  ← Now enabled! (no [disabled] attribute)

Key: The Save button will now show [cursor=pointer] and will NOT have [disabled] attribute.

Click Save:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"click","ref":"<save_button_ref>"}

Wait and verify:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"wait","timeMs":2000}

Take screenshot to confirm update.

Common mistakes to avoid:

Don't forget to select all (Cmd+A) before typing — otherwise you append to existing text!
Don't skip the wait after clicking — buttons need time to load.
Always take a fresh snapshot to get new ref numbers.

Add a New Note (新增註記 - Simple Steps for AI)

When to use: Contact has NO existing note. You will see ONLY a single button/link (no existing note text).

Step-by-step:

Open the conversation (click customer name in chat list). Wait 2 seconds for it to load.

Take snapshot to find the "add note" button:

browser action:"snapshot" profile:"openclaw" targetId:"<targetId>" refs:"aria" compact:true

Look for the pattern (Notes section with NO existing note):

- heading [level=3] [ref=eAAA]  ← Notes heading
- button [ref=eXXX]             ← Single "add note" button (THIS ONE!)

Key: You should see ONLY one button/link under the Notes heading, with NO paragraph element containing note text.

Click "add note" button:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"click","ref":"<button_ref>"}

Wait for the textbox:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"wait","timeMs":1500}

Take snapshot to find the textbox:

browser action:"snapshot" profile:"openclaw" targetId:"<targetId>" refs:"aria" compact:true

Look for:

- textbox [ref=eXXX]:              ← Input field (may show "Hidden Label")
  /placeholder: ...                 ← Placeholder text
- button [ref=eYYY]                 ← Cancel button
- button [disabled] [ref=eZZZ]     ← Save button (disabled at first)

Key: The textbox might be labeled "Hidden Label" or similar. Look for [disabled] attribute on the Save button.

Type the note text:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"type","ref":"<textbox_ref>","text":"Customer note text here"}

Wait for Save button to become enabled:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"wait","timeMs":500}

Take snapshot to find the enabled Save button:

browser action:"snapshot" profile:"openclaw" targetId:"<targetId>" refs:"aria" compact:true

Look for:

- button [ref=eXXX] [cursor=pointer]:  ← Now enabled (no [disabled] attribute!)

Click Save button:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"click","ref":"<save_button_ref>"}

Wait and verify:

browser action:"act" profile:"openclaw" targetId:"<targetId>" request:{"kind":"wait","timeMs":2000}

Take a screenshot to confirm the note appears in the sidebar.

Common mistakes to avoid:

Don't skip snapshots — you need the ref numbers!
Don't reuse old ref numbers — they change every snapshot.
Wait after clicking before taking next snapshot.

Delete a Note

Take a snapshot to find the Delete link (appears next to Edit link when a note exists).
Click the Delete link (the link that appears AFTER the "·" separator, next to the Edit link).
Confirm the deletion if prompted.

In snapshot:

- link [ref=eXXX]  ← Edit link
- text: "·"
- link [ref=eYYY]  ← Delete link (THIS ONE!)

Switch Between Pages

If multiple pages are configured:

Read config to see all available pages.

Navigate to the desired page URL:

browser action:"navigate" profile:"openclaw" targetId:"<targetId>" targetUrl:"<newPageUrl>"

Wait for the page to load.

Notes

Meta Business Suite sessions expire periodically; re-login may be required.
Facebook's DOM structure is complex and changes frequently. Prefer screenshots over DOM parsing when snapshots are unreliable.
The inbox URL format: https://business.facebook.com/latest/inbox/all/?&asset_id=<PAGE_ID>
Some pages may require specific permissions in Meta Business Suite.

When it activates

Automatically, when a task matches. Use when asked to check Facebook messages, reply to FB customers, or manage Facebook page inbox. Supports multiple pages with custom aliases.

About this skill

Automatic

Support pack

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 7/30/2026

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/meta-fb-inbox/SKILL.md

View on GitHub ↗

License MITImported 8/2/2026Version v1Bundled 3 files · scripts

Illustrative Want to see what meta-fb-inbox 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

Meta Facebook Inbox

Configuration

Config Format

Resolving a Page

Enter Facebook Inbox

Quick Check (for cron / automated checks)

Check Inbox Messages

Get URLs for Conversations (取得對話網址 - Simple!)

View Specific Conversation

Extract Messages with Images

Reply to a Message

Manage Labels

Remove a Label

Add a Label

Manage Notes (備註)

Quick Decision Guide (看這裡！)

Note Structure

Edit an Existing Note (編輯註記 - Simple Steps for AI)

Add a New Note (新增註記 - Simple Steps for AI)

Delete a Note

Switch Between Pages

Notes

When it activates

SkillSafety

Source

Related skills

More in the registry