Ready to supercharge your AI agents?

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

Get Started Free →

Smart Email Agent v3 — Gmail All-in-One · LeoYeAI | DecimalAI

Productivityby LeoYeAI

Smart Email Agent v3 — Gmail All-in-One

Un solo skill que reemplaza los 6 skills del pack original.

Never made a case worse

SkillScore

SkillScore73/100

Test results 73

Safety

3/3 safety checks passed

Code scan
Intent
Content

What it changes

0%73%of test cases passed +73 pts

Without skill0 of 22

With skill16 of 22

16 fixed0 passed either way6 still failing0 broke

Measured on gemini-3.6-flash7/27/2026v120 comparable

Track record

Agent runs: 0; 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 smart-email-agent
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("smart-email-agent")
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 smart-email-agent

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/smart-email-agent/SKILL.md https://app.decimal.ai/s/smart-email-agent/SKILL.md

Copy SKILL.md

Save it in your project, as

.claude/skills/smart-email-agent/SKILL.md

This gets SKILL.md only. The skill's 4 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 smart-email-agent

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("smart-email-agent")
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/smart-email-agent/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/smart-email-agent/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 4 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 smart-email-agent

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("smart-email-agent")
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/smart-email-agent/SKILL.md

Give it to your agent

from pathlib import Path
from agents import Agent, Runner

skill = Path(".agents/skills/smart-email-agent/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 4 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 smart-email-agent

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("smart-email-agent")
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/smart-email-agent/SKILL.md

Give it to your agent

from pathlib import Path
from openai import OpenAI

client = OpenAI()
skill = Path(".agents/skills/smart-email-agent/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 4 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 smart-email-agent

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("smart-email-agent")
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/smart-email-agent/SKILL.md

Give it to your agent

from pathlib import Path
import anthropic

client = anthropic.Anthropic()
skill = Path(".agents/skills/smart-email-agent/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 4 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 smart-email-agent

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("smart-email-agent")
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/smart-email-agent/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/smart-email-agent/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 4 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 smart-email-agent

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("smart-email-agent")
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/smart-email-agent/SKILL.md

Give it to your agent

from pathlib import Path
from langchain.agents import create_agent

skill = Path(".agents/skills/smart-email-agent/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 4 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 smart-email-agent

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("smart-email-agent")
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/smart-email-agent/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 4 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 smart-email-agent

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("smart-email-agent")
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/smart-email-agent/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/smart-email-agent/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 4 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 smart-email-agent

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("smart-email-agent")
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/smart-email-agent/SKILL.md

Give it to your agent

from pathlib import Path
from pydantic_ai import Agent

skill = Path(".agents/skills/smart-email-agent/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 4 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 smart-email-agent

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("smart-email-agent")
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/smart-email-agent/SKILL.md

Give it to your agent

from pathlib import Path
from deepagents import create_deep_agent

skill = Path(".agents/skills/smart-email-agent/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 4 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 smart-email-agent

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("smart-email-agent")
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/smart-email-agent/SKILL.md

Give it to your agent

import asyncio
from pathlib import Path
from agent_framework import Agent

skill = Path(".agents/skills/smart-email-agent/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 4 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 smart-email-agent

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("smart-email-agent")
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/smart-email-agent/SKILL.md

Give it to your agent

from pathlib import Path
from strands import Agent

skill = Path(".agents/skills/smart-email-agent/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 4 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 smart-email-agent

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/smart-email-agent/SKILL.md https://app.decimal.ai/s/smart-email-agent/SKILL.md

Copy SKILL.md

Save it in your project, as

.agents/skills/smart-email-agent/SKILL.md

This gets SKILL.md only. The skill's 4 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/smart-email-agent/SKILL.md

Read the summary as JSON

curl -fsSL https://app.decimal.ai/s/smart-email-agent.json

This gets SKILL.md only. The skill's 4 other files are not included.

Other ways to add it

Other setups docs

You don't need a DecimalAI account.

Version: v1, tested 7/27/2026
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

Smart Email Agent v3 — Gmail All-in-One

Un solo skill que reemplaza los 6 skills del pack original. Punto de entrada único para toda la gestión de correo.

Lazy loading activo: este skill se carga completo (~500 tokens). NO cargues los 6 skills individuales — sería 4.800 tokens desperdiciados.

PARTE 1 — LEER CORREOS (email-reader)

Herramienta principal: `gog` CLI

# Verificar prerequisitos
which gog || echo "Instalar: npm i -g gogcli  OR  brew install gogcli"
echo $GOG_ACCOUNT  # debe estar configurado

# Autenticar si es la primera vez
gog auth add $GOG_ACCOUNT

Comandos esenciales

# No leídos en inbox (acción por defecto cuando el usuario dice "revisa mi correo")
gog gmail search 'in:inbox is:unread' --max 5 --format minimal --json

# Buscar por criterio — usar sintaxis Gmail
gog gmail search 'from:[email protected] newer_than:3d' --max 10 --format minimal --json
gog gmail search 'subject:factura has:attachment' --max 10 --format minimal --json
gog gmail search 'in:spam is:unread' --max 20 --format minimal --json

# Leer correo completo
gog gmail get <message_id> --format full --json

# Leer hilo completo
gog gmail thread <thread_id> --format minimal --json

Operadores de búsqueda Gmail

from: to: subject: label: is:unread is:starred has:attachment newer_than:Nd older_than:Nd in:inbox in:sent in:spam in:trash filename:ext

Flujo estándar de lectura

Verificar gog en PATH y GOG_ACCOUNT configurado
Construir query desde la intención del usuario (preguntar si es ambiguo)
Ejecutar con --format minimal --json --max N
Parsear JSON → presentar lista formateada:

📬 5 correos no leídos:
1. De: Juan García <[email protected]> | Asunto: Propuesta Q2 | Hace 2h
   Vista previa: Hola, te mando el resumen de... | ID: msg_abc123

Ofrecer: leer completo / buscar más / actuar sobre el mensaje

Reglas de lectura

SIEMPRE usar --format minimal --json --max N (N=5 por defecto)
Nunca mostrar JSON crudo; nunca leer contenido completo sin pedirlo
Preservar IDs para acciones de seguimiento
Sin resultados → confirmar criterios, sugerir términos más amplios
Solo lectura — enviar/responder requiere la sección RESPONDER (Parte 4)
No guardar contenido de correos en MEMORY.md salvo que el usuario lo pida

Errores comunes

Error	Causa	Solución
`gog: command not found`	gog no instalado	`npm i -g gogcli` o `brew install gogcli`
`GOG_ACCOUNT not set`	Variable no configurada	Pedir al usuario su email Gmail
Token expirado	OAuth vencido	`gog auth add <email>`
API error 429	Rate limit	Esperar 60s, reintentar con backoff

PARTE 2 — ORGANIZAR (email-organizer)

Jerarquía corporativa Corp/ (azul)

Corp/
├── Interno/
│   ├── Management      ← Gerencia, Directores
│   ├── Tech & Ops      ← Desarrollo, Soporte, Operaciones
│   ├── Commercial      ← Ventas, Marketing
│   ├── Admin & HR      ← Jurídica, RRHH, Contabilidad
│   └── Team            ← Resto del equipo @empresa.com
├── Partners & Clientes/
│   └── [empresa]       ← Wolkvox, Masiv, Unisanitas, Nuva, etc.
├── Proveedores/
│   └── [proveedor]     ← Google, Microsoft, AWS, etc.
└── Sistema/
    ├── DMARC
    ├── Notificaciones
    ├── Alertas
    └── No-Reply

Comandos de organización con gog

# Crear etiqueta
gog gmail label create "Corp/Interno/Tech & Ops"

# Aplicar etiqueta a mensaje
gog gmail label apply <message_id> "Corp/Interno/Management"

# Mover correo (quitar INBOX + aplicar etiqueta)
gog gmail modify <message_id> --add-label "Corp/Partners & Clientes/Wolkvox" --remove-label INBOX

# Archivar (quitar INBOX sin borrar)
gog gmail modify <message_id> --remove-label INBOX

# Mover a spam
gog gmail modify <message_id> --add-label SPAM --remove-label INBOX

# Mover a papelera
gog gmail trash <message_id>

# Operación batch (múltiples IDs)
gog gmail batch-modify --ids id1,id2,id3 --add-label "Corp/Sistema/No-Reply" --remove-label INBOX

Reglas de routing automático

Antes de llamar a la IA, aplicar estas reglas sin costo:

from_domain @empresa.com + from_name contiene [Linda, Rafael, Director] → Corp/Interno/Management
from_domain @empresa.com + from_name contiene [Tech, Dev, Soporte]      → Corp/Interno/Tech & Ops
from_domain @empresa.com                                                  → Corp/Interno/Team
from contiene noreply / no-reply / donotreply                            → Corp/Sistema/No-Reply
subject contiene DMARC / SPF / DKIM                                      → Corp/Sistema/DMARC
subject contiene alerta / alert / warning                                → Corp/Sistema/Alertas

Guardar y mantener estas reglas en corp_routing_rules.json.

Protocolo de confirmación

NUNCA ejecutar acciones destructivas sin confirmación explícita:

⚠️ Pendiente de confirmación:
   → Mover 22 correos a Corp/Sistema/No-Reply
   → Eliminar etiqueta "noreply" (ya vacía)
   Esto NO borra correos, solo reorganiza etiquetas.
   ¿Confirmas? (sí/no)

PARTE 3 — ANALIZAR Y CLASIFICAR (email-analyzer)

Decisión de modelo ANTES de analizar

¿Tarea es clasificar / detectar spam / routing?
  → claude-haiku-4-5-20251001   (batch de 10-20 correos, ~$0.00009/correo)

¿Tarea es extraer tareas y fechas de correos importantes?
  → claude-haiku-4-5-20251001   (body[:800], ~$0.00015/correo)

¿El presupuesto está > 80% gastado?
  → forzar haiku para TODO, sin borradores automáticos

¿El presupuesto está > 95% gastado?
  → cero llamadas IA, solo reglas locales

Opus: PROHIBIDO para tareas de email. Sonnet: solo para borradores (ver Parte 4).

Pipeline de reducción de tokens (aplicar siempre)

# 1. Pre-filtro sin IA (resolver antes de gastar tokens)
#    - Dominio en corp_routing_rules.json → etiquetar directo
#    - from en known_spam_domains.txt → spam directo
#    - message_id ya en analysis_cache → reutilizar resultado
#    Objetivo: resolver 60-70% a costo $0.00

# 2. Recortar campos al mínimo necesario
CAMPOS = {
    'clasificacion': ['from', 'subject', 'snippet[:100]'],   # ~30 tokens
    'prioridad':     ['from', 'subject', 'body[:400]'],      # ~150 tokens
    'tareas_fechas': ['from', 'subject', 'body[:800]'],      # ~250 tokens
}

# 3. Limpiar texto
def limpiar(texto, limite):
    texto = re.sub(r'<[^>]+>', '', texto)             # quitar HTML
    texto = re.sub(r'https?://\S+', '[URL]', texto)   # comprimir URLs
    texto = re.sub(r'\s+', ' ', texto).strip()
    return texto[:limite]

# 4. Batch: NUNCA menos de 10 correos por llamada
#    Esperar hasta tener 10-20 correos pendientes
BATCH_MIN = 10
BATCH_MAX = 20

Prompt de análisis en batch (Haiku)

SYSTEM (idéntico siempre — para prompt caching):
Eres un clasificador de correos corporativos.
Analiza cada correo y devuelve SOLO JSON array. Sin texto extra.
Para cada ítem: {"idx":N,"corp_label":"...","categoria":"spam|importante|informativo|sistema|otro",
"prioridad":0-10,"es_spam":bool,"necesita_respuesta":bool,
"tiene_phishing":bool,"tareas":[],"fecha_limite":"ISO o null","razon":"máx 10 palabras"}

USER: Analiza: [JSON array de hasta 20 correos con from+subject+snippet[:100]]

Presentación de resultados

🤖 Análisis — 47 correos procesados
⚡ Sin IA (pre-filtro): 31  (66%) → $0.000
🧠 Con Haiku (2 batches): 16     → $0.006

📊 Resultado:
  🔵 Corp/Interno/Management:   2  (prioridad alta)
  🔵 Corp/Partners & Clientes:  8
  🔵 Corp/Sistema/No-Reply:    14
  🗑️  Spam:                    12
  ⚠️  Phishing detectado:        1  → ALERTA
  📋 Con tareas pendientes:      4

Críticos:
  [10/10] [email protected] — "Aprobación contrato urgente"
          Tarea: confirmar antes del viernes

PARTE 4 — RESPONDER Y REDACTAR (email-responder)

Cuándo usar Sonnet vs Haiku para borradores

Prioridad >= 8 → claude-sonnet-4-6       (calidad importa)
Prioridad 5-7  → claude-haiku-4-5-20251001  (suficiente, más barato)
Prioridad < 5  → NO generar borrador automático

Máximo 3 borradores por sesión cuando presupuesto < 60%. Máximo 1 borrador por sesión cuando presupuesto 60-80%. Cero borradores automáticos cuando presupuesto > 80%.

Flujo de respuesta

# 1. Leer el hilo completo
gog gmail thread <thread_id> --format minimal --json

# 2. Preparar contexto recortado para la IA
#    Solo: from + subject + body[:600] del último mensaje + resumen del hilo anterior

Prompt de generación de borrador

Redacta una respuesta profesional y concisa (máx 150 palabras).
Solo el cuerpo del mensaje, sin asunto ni encabezados.
Tono: profesional pero cercano.
Firma: [NOMBRE_USUARIO]

Hilo: [RESUMEN + ÚLTIMO MENSAJE RECORTADO]

Presentar borrador al usuario

✍️ Borrador para: [email protected]
   Re: Propuesta Q2 2026
────────────────────────
Hola Juan,

Gracias por el resumen. Me parece viable la dirección propuesta.
¿Podemos agendar una llamada esta semana?

Saludos,
[Tu nombre]
────────────────────────
[1] Guardar borrador   [2] Editar   [3] Enviar ahora   [4] Descartar

Enviar con gog

# Guardar como borrador
gog gmail draft create --to "[email protected]" \
  --subject "Re: Propuesta Q2 2026" \
  --body "Hola Juan,..." \
  --reply-to <message_id>

# Enviar borrador guardado
gog gmail draft send <draft_id>

# Enviar directamente (SIEMPRE pedir confirmación antes)
gog gmail send --to "[email protected]" --subject "..." --body "..."

Templates de respuesta rápida

acuse_recibo:      "Recibido, te respondo a la brevedad."
confirmar_reunion: "Confirmado para [fecha/hora]. Hasta entonces."
solicitar_info:    "Necesito más información sobre X para proceder."
ausencia:          "Estoy fuera hasta [fecha]. Respondo a mi regreso."

Follow-ups automáticos

# Buscar correos enviados sin respuesta en últimos 5 días
gog gmail search 'in:sent newer_than:5d' --max 20 --format minimal --json
# Cruzar con INBOX para detectar cuáles no tienen respuesta

PARTE 5 — AUTOMATIZAR (email-scheduler)

Heartbeat optimizado: 55 minutos

Por qué 55 min: el caché de Anthropic expira a los 60 minutos. Con heartbeat de 55 min, el agente mantiene el caché caliente → cada mensaje paga cache-read en lugar de cache-write (3.75x más barato).

// ~/.openclaw/openclaw.json
{
  "agents": {
    "email-assistant": {
      "heartbeat": { "every": "55m" },
      "model": "anthropic/claude-haiku-4-5-20251001"
    }
  }
}

Cron jobs recomendados

{
  "cron": {
    "jobs": [
      {
        "id": "email-priority-check",
        "schedule": "*/55 * * * *",
        "description": "Revisar correos importantes — modelo Haiku",
        "message": "Revisa inbox no leídos. Si hay prioridad >= 8, notifícame.",
        "model": "anthropic/claude-haiku-4-5-20251001",
        "enabled": true
      },
      {
        "id": "email-spam-cleanup",
        "schedule": "0 8 * * *",
        "description": "Limpieza diaria de spam — solo reglas locales, costo $0",
        "message": "Aplica reglas locales de spam. Sin llamadas IA.",
        "model": "anthropic/claude-haiku-4-5-20251001",
        "enabled": true
      },
      {
        "id": "email-weekly-report",
        "schedule": "0 9 * * MON",
        "description": "Informe semanal",
        "message": "Genera informe semanal de correos con email-reporter.",
        "model": "anthropic/claude-haiku-4-5-20251001",
        "enabled": true
      }
    ]
  }
}

Regla de oro para crons: SIEMPRE especificar claude-haiku-4-5-20251001. Usar Opus para un cron de 10 tareas/día = $17.70/mes extra innecesario.

HEARTBEAT.md de correo

## Email Heartbeat — Modelo: claude-haiku-4-5-20251001

### Check de correos (cada 55 min)
1. gog gmail search 'in:inbox is:unread' --max 10 --format minimal --json
2. Aplicar reglas locales de corp_routing_rules.json
3. Si hay correo con keywords urgente/crítico/emergencia → notificar
4. Si no hay urgentes → HEARTBEAT_OK (silencio)

### Check de spam (cada 2h, sin IA)
1. gog gmail search 'in:spam is:unread' --max 50 --format minimal --json
2. Aplicar known_spam_domains.txt → mover a trash directo
3. Sin llamadas a IA

Quiet hours: 23:00–07:00 → HEARTBEAT_OK automático

Gmail Push Notifications (tiempo real)

# Configurar webhook Pub/Sub
python3 scripts/setup_pubsub.py --topic "email-agent-notifications"

# El webhook dispara cuando llega un correo nuevo:
openclaw message "Nuevo correo. Revisa con gog y notifícame si es importante."

Activar automatización completa

Usuario: "Activa el agente de correo en modo automático"
Agente:
  1. Verificar: gog auth status
  2. Crear cron jobs recomendados (ver arriba)
  3. Copiar HEARTBEAT.md al workspace
  4. Preguntar: ¿activar Gmail Push para tiempo real?
  5. Confirmar canal de notificaciones (NOTIFY_CHANNEL)
  6. "✅ Agente activado. Reviso cada 55 min. Te aviso si hay algo importante."

PARTE 6 — INFORMES Y ESTADÍSTICAS (email-reporter)

Tipos de informe

# Resumen del día (al final de sesión — SIEMPRE mostrar)
# Ver sección "Resumen de costos" más abajo

# Estadísticas de spam
gog gmail search 'in:trash newer_than:30d' --max 100 --format minimal --json
# Parsear y agrupar por dominio remitente

# Tareas pendientes en correos
# Consultar analysis_cache donde tareas[] no está vacío y sin respuesta

# Log de prompts IA detectados
cat .learnings/PROMPTS_DETECTADOS.md

# Historial de acciones del agente
cat email_audit.log | tail -50

# Deshacer última acción
gog gmail modify <ids_from_audit_log> --remove-label TRASH --add-label INBOX

Resumen de costos — mostrar al cerrar cada sesión

💰 Sesión de hoy
  Correos procesados:    47
  ├─ Sin IA (reglas):    31  (66%) → $0.000
  ├─ Haiku (2 batches):  14       → $0.005
  └─ Sonnet (borradores): 2       → $0.005

  Tokens consumidos: 5.090
  Costo sesión:     $0.010
  Ahorro estimado:  $0.040 (80% vs. sin optimizar)

  Presupuesto mes:  $X.XX gastado / $Y.YY total (N%)
  Proyección mes:   $Z.ZZ

PARTE 7 — MOTOR DE AUTO-MEJORA

Cuándo capturar un aprendizaje

Evento	Archivo	ID
Usuario corrige clasificación	`.learnings/LEARNINGS.md`	`LRN-YYYYMMDD-NNN`
Costo sesión > 2x el promedio	`.learnings/LEARNINGS.md`	`LRN-YYYYMMDD-NNN`
Error de API (rate limit, auth)	`.learnings/ERRORS.md`	`ERR-YYYYMMDD-XXX`
Optimización reduce costos >10%	`.learnings/LEARNINGS.md`	`LRN-YYYYMMDD-NNN`
Remitente recurrente sin regla	`.learnings/LEARNINGS.md`	`LRN-YYYYMMDD-NNN`

Formato

## [LRN-YYYYMMDD-NNN] <tipo>
**Logged**: ISO timestamp
**Priority**: low | medium | high
**Status**: pending | applied | promoted
**Area**: cost | classification | routing | api | spam | drafts

### Summary
Una línea con el aprendizaje y su impacto.

### Details
Qué pasó. Qué se asumía vs. qué era verdad.

### Action
Cambio concreto: qué archivo editar, qué valor cambiar.

### Impact
Ahorro estimado: $X/mes | Tokens -N%
---

Ciclo al cerrar sesión

1. Revisar .learnings/ con Status=pending
2. ¿Learnings con Impact > $0.01/mes?
   → Proponer: "Aprendí que X. ¿Lo aplico a las reglas?"
   → Si acepta → editar corp_routing_rules.json → Status=applied
3. ¿3+ learnings del mismo dominio/patrón?
   → Promover a regla permanente sin preguntar
   → Status=promoted
4. Reportar: "Apliqué N mejoras. Ahorro estimado: $X/mes"

Efecto compuesto del aprendizaje

Mes 1: ~$0.50/mes  (0 reglas)
Mes 2: ~$0.30/mes  (10 reglas aprendidas)
Mes 3: ~$0.18/mes  (25 reglas)
Mes 6: ~$0.10/mes  (60+ reglas)

PARTE 8 — CONTROL DE PRESUPUESTO

Cuatro modos operativos

% gastado	Modo	Restricciones
0–59%	Normal ✅	Todo habilitado
60–79%	Ahorro leve 🟡	Avisar. Máx 2 borradores/sesión
80–94%	Ahorro fuerte 🟠	Solo Haiku. Sin borradores auto. Batch obligatorio ≥20
95–100%	Emergencia 🔴	Cero IA. Solo `gog` + reglas locales

Diagnósticos con token-optimizer (si está instalado)

/context list    # qué archivos consumen tokens ahora
/usage full      # tokens + costo por respuesta
/usage cost      # resumen acumulado de sesión
/status          # modelo activo, % contexto

python3 scripts/token_tracker.py check   # estado del budget diario
python3 scripts/model_router.py "analizar correos nuevos"  # qué modelo usar

Referencias — leer cuando necesites más detalle

references/cost-optimization.md — Técnicas avanzadas: prompt caching, deduplicación semántica, modo emergencia
references/learning-patterns.md — Patrones de auto-mejora y ciclo de vida de learnings
hooks/openclaw-handler.js — Inyecta estado de presupuesto + modo activo al inicio de sesión
scripts/init_orchestrator.py — Setup inicial: verifica gog, crea budget_tracker, SKILLS_INDEX
assets/HEARTBEAT.email.md — Plantilla lista para copiar al workspace

About this skill

Automatic

Personal Assistant packemailgmailgog

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/emailagy/SKILL.md

View on GitHub ↗

License MITImported 8/2/2026Version v1Bundled 4 files · scripts

Illustrative Want to see what smart-email-agent 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

Smart Email Agent v3 — Gmail All-in-One

PARTE 1 — LEER CORREOS (email-reader)

Herramienta principal: gog CLI

Comandos esenciales

Operadores de búsqueda Gmail

Flujo estándar de lectura

Reglas de lectura

Errores comunes

PARTE 2 — ORGANIZAR (email-organizer)

Jerarquía corporativa Corp/ (azul)

Comandos de organización con gog

Reglas de routing automático

Protocolo de confirmación

PARTE 3 — ANALIZAR Y CLASIFICAR (email-analyzer)

Decisión de modelo ANTES de analizar

Pipeline de reducción de tokens (aplicar siempre)

Prompt de análisis en batch (Haiku)

Presentación de resultados

PARTE 4 — RESPONDER Y REDACTAR (email-responder)

Cuándo usar Sonnet vs Haiku para borradores

Flujo de respuesta

Prompt de generación de borrador

Presentar borrador al usuario

Enviar con gog

Templates de respuesta rápida

Follow-ups automáticos

PARTE 5 — AUTOMATIZAR (email-scheduler)

Heartbeat optimizado: 55 minutos

Cron jobs recomendados

HEARTBEAT.md de correo

Gmail Push Notifications (tiempo real)

Activar automatización completa

PARTE 6 — INFORMES Y ESTADÍSTICAS (email-reporter)

Tipos de informe

Resumen de costos — mostrar al cerrar cada sesión

PARTE 7 — MOTOR DE AUTO-MEJORA

Cuándo capturar un aprendizaje

Formato

Ciclo al cerrar sesión

Efecto compuesto del aprendizaje

PARTE 8 — CONTROL DE PRESUPUESTO

Cuatro modos operativos

Diagnósticos con token-optimizer (si está instalado)

Referencias — leer cuando necesites más detalle

SkillSafety

Source

Related skills

More in the registry

Herramienta principal: `gog` CLI