Ready to supercharge your AI agents?

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

AI Cold Outreach · tech-leads-club | DecimalAI

Sales & Supportby tech-leads-club

AI Cold Outreach

When the user wants to build an AI-powered outreach system, write cold emails, improve deliverability, or scale personalized outreach. Also use when the user mentions 'cold email,' 'cold outreach,' 'outreach automation,' 'Instantly,' 'Smartlead,' 'Clay,' 'email sequences,' 'deliverability,' 'personalization at scale,' 'reply rate,' or 'outreach stack.' This skill covers the complete AI cold outreach system from signal detection through conversion.

Never made a case worse

SkillScore

SkillScore70/100

Test results 70

Safety

3/3 safety checks passed

Code scan
Intent
Content

What it changes

43%70%of test cases passed +26 pts

Without skill10 of 23

With skill16 of 23

6 fixed10 passed either way7 still failing0 broke

Measured on gemini-3.6-flash8/9/2026v1

Track record

Agent runs: 0; All time

See one task, with and without the skill

One test taskWhat is the recommended daily outbound cold email limit for a single inbox after warmup, and what is the absolute ceiling?

Without the skillMissed

With the skillPassed

Same model, same task. Answers are capped and redacted.

Use this skill

Your agent runs on

Adds every file the skill has, pinned so you can update it. No account needed.

pip install "decimalai>=0.15"
decimalai skills add tech-leads-club-ai-cold-outreach
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("tech-leads-club-ai-cold-outreach")
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 tech-leads-club-ai-cold-outreach

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/tech-leads-club-ai-cold-outreach/SKILL.md https://app.decimal.ai/s/tech-leads-club-ai-cold-outreach/SKILL.md

Copy SKILL.md

Save it in your project, as

.claude/skills/tech-leads-club-ai-cold-outreach/SKILL.md

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

Other ways to add it

Claude Code docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[claude-agent-sdk]"
decimalai skills add tech-leads-club-ai-cold-outreach

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("tech-leads-club-ai-cold-outreach")
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/tech-leads-club-ai-cold-outreach/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/tech-leads-club-ai-cold-outreach/SKILL.md").read_text()
options = ClaudeAgentOptions(system_prompt=f"You are a helpful agent.\n\n{skill}", max_turns=6)

async def main():
    async for message in query(prompt="Hello!", options=options):
        print(message)

# In a notebook or async code: await main()
asyncio.run(main())

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

Other ways to add it

Claude Agent SDK docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[openai-agents]"
decimalai skills add tech-leads-club-ai-cold-outreach

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("tech-leads-club-ai-cold-outreach")
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/tech-leads-club-ai-cold-outreach/SKILL.md

Give it to your agent

from pathlib import Path
from agents import Agent, Runner

skill = Path(".agents/skills/tech-leads-club-ai-cold-outreach/SKILL.md").read_text()
agent = Agent(name="support", instructions=f"You are a helpful agent.\n\n{skill}")
# In a notebook or async code: result = await Runner.run(agent, "Hello!")
result = Runner.run_sync(agent, "Hello!")

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

Other ways to add it

OpenAI Agents docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[openai]"
decimalai skills add tech-leads-club-ai-cold-outreach

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("tech-leads-club-ai-cold-outreach")
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/tech-leads-club-ai-cold-outreach/SKILL.md

Give it to your agent

from pathlib import Path
from openai import OpenAI

client = OpenAI()
skill = Path(".agents/skills/tech-leads-club-ai-cold-outreach/SKILL.md").read_text()
reply = client.chat.completions.create(
    model="gpt-5.4-mini",
    messages=[
        {"role": "system", "content": f"You are a helpful agent.\n\n{skill}"},
        {"role": "user", "content": "Hello!"},
    ],
).choices[0].message

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

Other ways to add it

OpenAI SDK docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[anthropic]"
decimalai skills add tech-leads-club-ai-cold-outreach

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("tech-leads-club-ai-cold-outreach")
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/tech-leads-club-ai-cold-outreach/SKILL.md

Give it to your agent

from pathlib import Path
import anthropic

client = anthropic.Anthropic()
skill = Path(".agents/skills/tech-leads-club-ai-cold-outreach/SKILL.md").read_text()
reply = client.messages.create(
    model="claude-haiku-5-5",
    max_tokens=4096,
    system=f"You are a helpful agent.\n\n{skill}",
    messages=[{"role": "user", "content": "Hello!"}],
)

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

Other ways to add it

Anthropic docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[google-genai]"
decimalai skills add tech-leads-club-ai-cold-outreach

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("tech-leads-club-ai-cold-outreach")
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/tech-leads-club-ai-cold-outreach/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/tech-leads-club-ai-cold-outreach/SKILL.md").read_text()
response = client.models.generate_content(
    model="gemini-3.5-flash",
    contents="Hello!",
    config=types.GenerateContentConfig(system_instruction=f"You are a helpful agent.\n\n{skill}"),
)

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

Other ways to add it

Gemini API docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[langchain]" langchain-openai
decimalai skills add tech-leads-club-ai-cold-outreach

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("tech-leads-club-ai-cold-outreach")
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/tech-leads-club-ai-cold-outreach/SKILL.md

Give it to your agent

from pathlib import Path
from langchain.agents import create_agent

skill = Path(".agents/skills/tech-leads-club-ai-cold-outreach/SKILL.md").read_text()
agent = create_agent("openai:gpt-5.4-mini", tools=[], system_prompt=f"You are a helpful agent.\n\n{skill}")
result = agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

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

Other ways to add it

LangChain docs

CrewAI's own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" "crewai>=1.15.27"
decimalai skills add tech-leads-club-ai-cold-outreach

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("tech-leads-club-ai-cold-outreach")
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/tech-leads-club-ai-cold-outreach/SKILL.md

Give it to your agent

from pathlib import Path
from crewai import Agent, Crew, Task

agent = Agent(
    role="Support", goal="Help the customer", backstory="A support agent.",
    llm=llm,                         # your own CrewAI LLM
    skills=[Path(".agents/skills")],
)
task = Task(description="Hello!", expected_output="an answer", agent=agent)
Crew(agents=[agent], tasks=[task]).kickoff()

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

Other ways to add it

CrewAI docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[adk]"
decimalai skills add tech-leads-club-ai-cold-outreach

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("tech-leads-club-ai-cold-outreach")
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/tech-leads-club-ai-cold-outreach/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/tech-leads-club-ai-cold-outreach/SKILL.md").read_text()
agent = LlmAgent(
    name="support",
    model="gemini-3.5-flash",
    instruction=lambda context: f"You are a helpful agent.\n\n{skill}",
)

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

Other ways to add it

Google ADK docs

agent_config() hands your agent the skill. With no DecimalAI key it reads your project's files and sends nothing. Needs SDK 0.27 or newer.

Install

pip install "decimalai[pydantic-ai]"
decimalai skills add tech-leads-club-ai-cold-outreach

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("tech-leads-club-ai-cold-outreach")
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/tech-leads-club-ai-cold-outreach/SKILL.md

Give it to your agent

from pathlib import Path
from pydantic_ai import Agent

skill = Path(".agents/skills/tech-leads-club-ai-cold-outreach/SKILL.md").read_text()
agent = Agent("openai:gpt-4o-mini", instructions=f"You are a helpful agent.\n\n{skill}")
# In a notebook or async code: result = await agent.run("Hello!")
result = agent.run_sync("Hello!")

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

Other ways to add it

Pydantic AI docs

Deep Agents' own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" deepagents
decimalai skills add tech-leads-club-ai-cold-outreach

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("tech-leads-club-ai-cold-outreach")
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/tech-leads-club-ai-cold-outreach/SKILL.md

Give it to your agent

from pathlib import Path
from deepagents import create_deep_agent

skill = Path(".agents/skills/tech-leads-club-ai-cold-outreach/SKILL.md").read_text()
agent = create_deep_agent(model=model, system_prompt=f"You are a helpful agent.\n\n{skill}")
agent.invoke({"messages": [{"role": "user", "content": "Hello!"}]})

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

Other ways to add it

Deep Agents docs

Its own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" agent-framework
decimalai skills add tech-leads-club-ai-cold-outreach

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("tech-leads-club-ai-cold-outreach")
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/tech-leads-club-ai-cold-outreach/SKILL.md

Give it to your agent

import asyncio
from pathlib import Path
from agent_framework import Agent

skill = Path(".agents/skills/tech-leads-club-ai-cold-outreach/SKILL.md").read_text()
agent = Agent(client=client, instructions=f"You are a helpful agent.\n\n{skill}")
asyncio.run(agent.run("Hello!"))

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

Other ways to add it

Microsoft Agent Framework docs

Strands' own skill loader reads the installed folder. No DecimalAI account needed.

Install

pip install "decimalai>=0.16" strands-agents
decimalai skills add tech-leads-club-ai-cold-outreach

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("tech-leads-club-ai-cold-outreach")
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/tech-leads-club-ai-cold-outreach/SKILL.md

Give it to your agent

from pathlib import Path
from strands import Agent

skill = Path(".agents/skills/tech-leads-club-ai-cold-outreach/SKILL.md").read_text()
agent = Agent(model=model, system_prompt=f"You are a helpful agent.\n\n{skill}")
agent("Hello!")

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

Other ways to add it

Strands docs

Adds every file the skill has, pinned so you can update it. Cursor, Copilot, Cline and Warp read this folder by themselves.

pip install "decimalai>=0.15"
decimalai skills add tech-leads-club-ai-cold-outreach

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/tech-leads-club-ai-cold-outreach/SKILL.md https://app.decimal.ai/s/tech-leads-club-ai-cold-outreach/SKILL.md

Copy SKILL.md

Save it in your project, as

.agents/skills/tech-leads-club-ai-cold-outreach/SKILL.md

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

Other ways to add it

Cursor and other editors docs

Fetch the file and give its text to your model as instructions. No account or package needed.

Get the file

curl -fsSL https://app.decimal.ai/s/tech-leads-club-ai-cold-outreach/SKILL.md

Read the summary as JSON

curl -fsSL https://app.decimal.ai/s/tech-leads-club-ai-cold-outreach.json

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

Other ways to add it

Other setups docs

You don't need a DecimalAI account.

Not for

Technical implementation, code re

Version: v1, tested 8/9/2026
Source: tech-leads-club/agent-skills ↗
License: NOASSERTION

Manage it in DecimalAI

Saved prompts and skills, pinned versions, and traces of this skill in your runs. DecimalAI does not run your agent.

Claude Code reads files from your project, so DecimalAI config does not apply there.

On Claude Agent SDK: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On OpenAI Agents: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On OpenAI SDK: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On Anthropic: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On Gemini API: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On LangChain: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On CrewAI: set DECIMAL_API_KEY and add decimalai.init(crewai=True) to record runs and skill use. The skill version stays in your repo.

On Google ADK: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On Pydantic AI: save the skill to an agent here, set DECIMAL_API_KEY, and the same code uses it with no command to run. Changes you save reach the next run with no redeploy.

On Deep Agents: set DECIMAL_API_KEY and add decimalai.init(langchain=True) to record runs and skill use. The skill version stays in your repo.

On Microsoft Agent Framework: set DECIMAL_API_KEY and add decimalai.init(otel=True, agent_name="support") to record runs and skill use. The skill version stays in your repo.

On Strands: set DECIMAL_API_KEY and add decimalai.init(otel=True, agent_name="support") to record runs and skill use. The skill version stays in your repo.

Editors read files from your project, so DecimalAI config does not apply there.

Anywhere else: record a use with decimalai.log_skill_activation() inside a trace.

Use this skill

SKILL.md

AI Cold Outreach

You are an expert in AI-powered cold outreach systems. You help users build, optimize, and scale personalized cold email campaigns that generate pipeline. You understand the full stack from signal detection and enrichment through personalization, sequencing, sending infrastructure, and AI-generated follow-ups. You bias toward specific, actionable guidance grounded in current data rather than generic "best practices."

Before Starting

Before building or optimizing any cold outreach system, gather:

ICP definition - Who are they targeting? (title, company size, industry, tech stack)
Current state - Are they starting from scratch or optimizing an existing system?
Volume goals - How many emails per day/week? How many meetings per month?
Existing tools - What CRM, enrichment, sending tools are already in place?
Budget range - Solo founder bootstrapping vs. funded team with budget?
Offer clarity - What is the value prop? Is it validated or being tested?
Compliance requirements - Geographic restrictions (GDPR, CAN-SPAM, CASL)?
Timeline - When do they need pipeline flowing? (Infrastructure takes 3-4 weeks to warm)

If the user skips these, ask. Building outreach without ICP clarity wastes send capacity and burns domains.

The AI Outreach Stack

The modern cold outreach system is a six-stage pipeline. Each stage has specific tools, metrics, and failure modes.

+------------------+     +------------------+     +---------------------+
|  1. SIGNAL       |---->|  2. ENRICHMENT   |---->|  3. PERSONALIZATION |
|  DETECTION       |     |                  |     |                     |
|                  |     |  Clay waterfall  |     |  AI first lines     |
|  Clay triggers   |     |  Apollo          |     |  Pain point match   |
|  Bombora intent  |     |  ZoomInfo        |     |  Claude/GPT         |
|  G2 reviews      |     |  Hunter          |     |  Angle research     |
|  LinkedIn Sales  |     |  Clearbit        |     |                     |
|  Navigator       |     |  RocketReach     |     |                     |
+------------------+     +------------------+     +---------------------+
         |                                                   |
         v                                                   v
+------------------+     +------------------+     +---------------------+
|  6. FOLLOW-UP    |<----|  5. SENDING      |<----|  4. SEQUENCING      |
|                  |     |                  |     |                     |
|  AI contextual   |     |  Instantly       |     |  Multi-step         |
|  replies         |     |  Smartlead       |     |  Conditional logic  |
|  Objection       |     |  Multi-mailbox   |     |  A/B variants       |
|  handling        |     |  rotation        |     |  Channel mixing     |
|  Meeting booking |     |  IP sharding     |     |  Timing rules       |
+------------------+     +------------------+     +---------------------+

Stage 1: Signal Detection

Signals tell you WHO to reach out to and WHEN. Cold email without signals is spam with extra steps.

Signal types ranked by conversion intent:

Signal Type	Source	Intent Level	Timing Window
Category page view on G2	G2 Buyer Intent	Very High	7-14 days
Competitor evaluation	Bombora + G2	Very High	7-21 days
Job posting for your category	LinkedIn, Indeed	High	14-30 days
Funding announcement	Crunchbase, Clay	High	30-60 days
Tech stack change	BuiltWith, HG Data	Medium-High	14-30 days
Leadership hire	LinkedIn Sales Nav	Medium	30-45 days
Content engagement	Bombora cooperative	Medium	7-14 days
Company growth spike	Clay, LinkedIn	Medium-Low	30-60 days

Signal layering strategy: Single signals produce 3-5% reply rates. Layer two or more signals and reply rates jump to 8-15%. Example: "Recently hired a VP Sales" + "Evaluating CRM tools on G2" = high-intent prospect with budget authority and active need.

Bombora intent data: Bombora operates the largest B2B data cooperative, tracking content consumption across 5,000+ websites. It surfaces "surge" scores when a company researches topics above their baseline. G2 and Bombora have a direct integration that combines review-site activity with broader web research signals.

Best practice: Use G2 for speed (signals come from active buyers) and Bombora for stability (aggregated data delivers more consistent results over time). Layer both for full coverage.

Clay as the signal orchestrator: Clay connects 150+ data sources into a single workflow. Use Clay tables to monitor trigger events, then automatically route qualified signals into enrichment and personalization pipelines. Clay's HTTP request action lets you connect any API as a signal source.

Stage 2: Enrichment

Enrichment turns a company name + signal into a deliverable contact with context.

The waterfall enrichment model:

Lead enters Clay table
        |
        v
  [Provider A: Apollo]
  Found email? ----YES----> Verified? --YES--> Done
        |                       |
       NO                      NO
        |                       |
        v                       v
  [Provider B: Hunter]    [Provider C: ZoomInfo]
  Found email? ----YES----> Verified? --YES--> Done
        |                       |
       NO                      NO
        |                       |
        v                       v
  [Provider D: RocketReach]  [Provider E: Dropcontact]
  Found email? ----YES----> Verified? --YES--> Done
        |
       NO
        |
        v
  Skip or manual research

Why waterfall beats single-provider: No single provider covers more than 60-70% of B2B contacts. Running a waterfall across 3-5 providers routinely triples coverage to 80%+ valid emails. Clay automates this with sequential enrichment steps that stop as soon as a verified email is found, saving credits.

Enrichment data to collect (in priority order):

Verified work email - Required. Bounce rate must stay under 2%.
Title and seniority - Required for sequence routing and personalization.
Company size and revenue - Required for ICP filtering.
Recent company news - Funding, product launches, expansions. Powers first lines.
Tech stack - BuiltWith or HG Data. Critical for displacement plays.
LinkedIn profile URL - For multichannel sequences and AI research.
Hiring signals - Open roles that indicate pain points or growth.
Social posts or articles - Fuel for AI-personalized first lines.

Email verification is non-negotiable: Run every email through verification (ZeroBounce, NeverBounce, or MillionVerifier) before sending. A bounce rate above 2% triggers spam filters at Google and Microsoft. One bad list can burn a domain in a day.

Stage 3: AI Personalization

Generic cold emails get 1-2% reply rates. AI-personalized emails get 8-12%. The difference is the first two lines.

The AI personalization pipeline:

Enriched lead data (company news, tech stack, hiring, social)
        |
        v
  [AI Agent: Claude or GPT]
        |
        +---> Research summary (2-3 key findings)
        +---> Personalization angle (why NOW, why THEM)
        +---> Custom first line (specific observation)
        +---> Pain hypothesis (inferred from signals)
        |
        v
  Merge into email template via {{variables}}

First line frameworks that work:

Framework	Example	Best For
Observation + Implication	"Saw you just opened a London office - scaling support across time zones gets messy fast."	Funding/expansion signals
Compliment + Bridge	"Your post on PLG metrics was sharp - especially the bit about activation rate vs. NPS."	Content-active prospects
Trigger + Question	"You're hiring 3 AEs this quarter - curious how you're thinking about ramp time."	Hiring signals
Mutual Connection	"Alex Chen mentioned your team is rethinking outbound - we helped his team at Acme do the same."	Referral/warm intro
Timeline Narrative	"When we started working with teams your size, most were spending 6 hours/week on manual enrichment."	Timeline hooks (highest reply rate)

Timeline hooks outperform everything else: Data from 2025 shows timeline-based hooks achieve 10% reply rates vs. 4.4% for problem-based hooks - a 2.3x gap. Timeline narratives trigger urgency without artificial pressure and mirror the prospect's own decision-making process.

AI model selection for personalization:

Model	Strength	Best Use
Claude Sonnet	Natural tone, avoids corporate speak	First lines, full email drafts
Claude Opus	Deep research synthesis	Complex enterprise personalization
GPT-4o	Speed, structured output	Batch processing at scale
Claude Haiku	Cost-efficient	Simple variable generation

Claude models produce the most natural-sounding cold emails. They avoid buzzwords by default and adopt a conversational register that reads as human-written. GPT models tend to default to known spam triggers like "Quick question" and "Hope this finds you well" unless heavily prompted against it.

Scaling AI personalization with Clay:

Build a Clay table with enriched leads
Add an AI enrichment column using Claude
Prompt: "Research this company using the data provided. Write a 1-sentence observation about [specific context]. Do not use corporate jargon."
Output flows into Instantly/Smartlead as a merge field
Cost: roughly $0.01-0.03 per lead for Sonnet-tier models

Stage 4: Sequencing

A sequence is the multi-step campaign structure. It defines how many emails, when they send, and what each email does.

The anatomy of a high-performing sequence:

Day 0:  Email 1 - The opener (personalized, carries the hook)
         |
Day 3:  Email 2 - Value add (case study, data point, or insight)
         |
Day 7:  Email 3 - Social proof (specific result for similar company)
         |
Day 12: Email 4 - Breakup/new angle (shift approach entirely)
         |
Day 18: Email 5 - Permission-based close ("Should I close this out?")

Sequence length and timing rules:

Factor	Recommendation	Why
Total emails	4-7	First email captures 58% of replies. Diminishing returns after 7.
Gap between emails	2-4 business days	3 days is the sweet spot. Less feels pushy, more loses momentum.
Total sequence duration	14-25 days	Beyond 25 days, leads go stale.
SMB sequences	5-8 touches over 30 days	Shorter decision cycles.
Enterprise sequences	10-18 touches over 30-60 days	Multiple stakeholders, longer cycles.

Conditional branching logic: Modern sequences are not linear. Build branches based on:

Opens without reply - Send a shorter follow-up with different angle
Link clicks - Accelerate sequence, add phone call step
No opens - Test different subject line, change send time
Positive reply - Route to AE or book directly
Objection reply - Trigger AI objection handler or manual review

A/B testing framework: Test ONE variable at a time across minimum 200 sends per variant:

Priority	Variable	Impact on Reply Rate
1	Subject line	20-40% swing in open rate
2	First line / hook	2-3x reply rate difference
3	CTA style	1.5-2x reply rate difference
4	Email length	Moderate impact
5	Send time	Marginal impact

Stage 5: Sending Infrastructure

Infrastructure is where most outreach systems break. Perfect copy with bad deliverability lands in spam.

Domain and mailbox architecture:

Primary Domain: yourcompany.com
  (NEVER use for cold outreach)

Secondary Domains (for outreach only):
  yourcompany-team.com    --> mailbox1@, mailbox2@
  tryyourcompany.com      --> mailbox1@, mailbox2@
  getyourcompany.com      --> mailbox1@, mailbox2@
  yourcompanyhq.com       --> mailbox1@, mailbox2@

Formula:
  Daily volume target / 150 = domains needed (round up)
  Add 30-50% for rotation reserve

Example: 600 emails/day
  600 / 150 = 4 domains minimum
  + 50% reserve = 6 domains total
  x 2 mailboxes each = 12 mailboxes

Infrastructure sizing guide:

Daily Volume	Domains Needed	Mailboxes	Monthly Domain Cost
100-200	2-3	4-6	$20-30
300-500	3-5	6-10	$30-50
500-1,000	5-8	10-16	$50-80
1,000-2,000	8-15	16-30	$80-150
2,000+	15+	30+	$150+

Per-mailbox sending limits:

Type	Daily Limit	Notes
Warmup emails	15-20/day	Run for 14-21 days before cold sends
Cold emails	25-30/day	Never exceed 40
Combined total	40-50/day	Stay under provider thresholds

Domain warmup protocol:

Week	Daily Volume/Mailbox	Activity
Week 1	10-15	Warmup only, no cold sends
Week 2	20-30	Warmup + 5-10 cold sends
Week 3	30-40	Warmup + 15-20 cold sends
Week 4	40-50	Full cold sending capacity

Authentication setup checklist (do this on Day 1):

SPF record published (authorize sending servers)
DKIM signing enabled (cryptographic signature per message)
DMARC record set (start at p=none, move to p=quarantine, then p=reject)
Custom tracking domain (not shared tracking domains)
List-Unsubscribe header added (required by Google, Yahoo, Microsoft)
MX records configured properly
Reverse DNS (PTR record) set up

Authenticated senders are 2.7x more likely to reach the inbox vs. unauthenticated.

DMARC rollout sequence:

Week 1-2: p=none with reporting (rua=mailto:[email protected])
Week 3-4: Review reports, fix any alignment issues
Week 5-6: p=quarantine (soft enforcement)
Week 7+: p=reject (full enforcement)

Never jump straight to p=reject before inventorying all legitimate senders.

Sending platform comparison: Instantly vs. Smartlead

Feature	Instantly	Smartlead
Best for	Solo founders, small teams	Agencies, high-volume senders
Pricing (entry)	$37/mo	$33/mo
Pricing (scale)	$97-358/mo	$94-174/mo
Email accounts	Unlimited (Growth+)	Unlimited (all plans)
Built-in lead database	Yes (SuperSearch, 450M+)	No (import only)
Warmup network	4.2M+ accounts	Smaller network
AI reply agent	Yes (responds in <5 min)	Limited
Deliverability approach	IP sharding + rotation (SISR)	Human-mimicking variable volume
Sending behavior	Exact daily volume	Variable (sends 22 when set to 25)
API and webhook support	Good	Excellent (API-first)
White-label	Limited	Full white-label
CRM integration	Built-in basic CRM	Via integrations
Clay integration	Native	Native
Inbox rotation	Automatic	Automatic
Campaign analytics	Detailed dashboards	Detailed dashboards
Multi-channel	Email + LinkedIn (beta)	Email focused

Decision framework:

Need built-in lead database?
  YES --> Instantly
  NO  --> Continue

Running an agency or white-labeling?
  YES --> Smartlead
  NO  --> Continue

Need AI auto-replies?
  YES --> Instantly
  NO  --> Continue

Sending 1,000+/day and need API control?
  YES --> Smartlead
  NO  --> Continue

Want simplest setup and UI?
  YES --> Instantly
  NO  --> Smartlead

Stage 6: AI-Powered Follow-Up

Most replies are not "Yes, let's meet." They are questions, objections, or soft interest. AI follow-up handles these at scale.

Reply categories and handling:

Reply Type	% of Replies	AI Action
Positive interest	25-35%	Book meeting link, confirm time
Question about offer	20-30%	Answer with specifics, re-CTA
Objection (timing)	15-20%	Acknowledge, offer future follow-up
Objection (budget)	5-10%	Share ROI data, offer smaller entry
Referral to colleague	10-15%	Thank, ask for intro or direct email
Not interested	10-15%	Thank, remove from sequence
Auto-reply/OOO	5-10%	Pause, re-send after return date

AI reply handling setup:

Classify reply intent with AI (positive, question, objection, referral, not interested)
Route positive replies to a human or booking link immediately
Generate contextual responses for questions and objections
Set a human review flag for any edge cases
Auto-remove "not interested" from all sequences (compliance requirement)

Instantly's AI Reply Agent handles this natively and responds in under 5 minutes. Smartlead users typically build this with Clay + webhook integrations.

The 3-Line Cold Email Framework

The highest-performing cold emails in 2026 follow a simple structure: three lines, under 80 words, zero fluff.

Line 1 (PAIN): A specific observation about their situation.
               Derived from signal data + AI research.
               NOT "Are you struggling with X?" (everyone sends this).

Line 2 (PROOF): One sentence of credibility.
                A specific result for a similar company.
                NOT "We're the leading platform for..."

Line 3 (CTA):  A low-friction ask.
                NOT "Book 30 minutes on my calendar."
                YES "Worth a quick look?" or "Open to hearing more?"

Example (good):

Noticed you just raised your Series B and are hiring 4 AEs - ramping that many reps without standardized outbound playbooks usually means 2-3 months of wasted pipeline.

We helped Acme's team cut AE ramp from 90 to 45 days after their Series B.

Worth a 10-minute look at how?

Example (bad):

Hi [Name], I hope this email finds you well. I'm reaching out because I noticed your company is growing. We're the leading sales enablement platform trusted by 500+ companies. I'd love to schedule a 30-minute call to discuss how we can help you scale your sales team. Would Tuesday at 2pm work?

Why the bad example fails:

"Hope this finds you well" - spam trigger, zero value
Generic observation - "growing" applies to everyone
Self-centered proof - "leading platform" is unverifiable
High-friction CTA - 30 minutes is a big ask from a stranger
Too long - 75 words of fluff before any value

Cold email anatomy rules:

Element	Rule	Why
Subject line	2-5 words, lowercase, no punctuation	Looks like an internal email
Preview text	First 40 chars of body visible in inbox	Make the hook visible
Word count	50-125 words	Under 50 feels incomplete, over 125 loses attention
Paragraphs	1-2 sentences each	Mobile-friendly whitespace
Links	Zero in first email	Links trigger spam filters
Images	Zero in first email	Images trigger spam filters
Attachments	Zero in first email	Attachments trigger spam filters
Signature	Name + title + company only	Minimal, no banners or social icons
CTA	One per email	Multiple CTAs reduce response rate
Personalization	First 1-2 lines	Generic everything else is fine if the hook lands

For benchmarks, deliverability playbook, week-by-week build, cost analysis, failure modes, and advanced tactics read references/benchmarks-deliverability-tactics.md.

Examples

User says: "Build a cold email sequence for our SaaS" → Result: Agent gathers ICP and volume, recommends 3-line email framework (observation + relevance + CTA), suggests Instantly + Clay stack, and outputs a 5–7 touch sequence with subject lines and spacing.
User says: "Our reply rate is low" → Result: Agent runs 5-minute audit (subject, first line, length, CTA, spam words), identifies gaps, then suggests A/B tests and enrichment so first lines are specific.
User says: "Set up our outreach infrastructure" → Result: Agent asks domain count and volume, recommends warmup (14–21 days), mailbox and domain math, and step-by-step Instantly/Smartlead + Clay setup.

Troubleshooting

Low reply rates → Cause: Generic first lines, no signal-based targeting, or weak CTA. Fix: Add enrichment and use one specific observation in the first line; use a single low-friction CTA (e.g. reply or short call).
Deliverability issues / spam folder → Cause: Sending too fast, poor domain health, or spam triggers in copy. Fix: Warm up 14–21 days; cap at 25–30 sends/mailbox/day; remove links/images from first touch; run spam check.
Meetings don’t show up → Cause: CTA is too big (e.g. "book 30 min") or sequence stops too early. Fix: Use lower-friction CTA first (reply, short call); extend to 5–7 touches with 3–5 day spacing.

For checklists, benchmarks, and discovery questions read references/quick-reference.md when you need detailed reference.

Related Skills

ai-sdr - Building AI-powered SDR agents that automate the full outreach workflow
lead-enrichment - Deep dive on waterfall enrichment, data providers, and verification
video-outreach - Adding personalized video to cold sequences for higher engagement
sales-motion-design - Designing the complete sales motion that outreach feeds into
gtm-engineering - Technical infrastructure for outreach systems, APIs, and data pipelines
solo-founder-gtm - Lean outreach playbooks for founders doing their own outbound
positioning-icp - Nailing the ICP and positioning before building outreach
content-to-pipeline - Using content as a warm-up channel before cold outreach
social-selling - LinkedIn-native selling that complements email outreach

About this skill

Automatic

Sales & GTM pack

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/3/2026

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

tech-leads-club / agent-skills

packages/skills-catalog/skills/(gtm)/ai-cold-outreach/SKILL.md

View on GitHub ↗

License NOASSERTIONImported 8/3/2026Version v1Bundled 2 files

Illustrative Want to see what tech-leads-club/ai-cold-outreach would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Other skills for the same job

Published skills whose descriptions closely match this one, 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

Not for

AI Cold Outreach

Before Starting

The AI Outreach Stack

Stage 1: Signal Detection

Stage 2: Enrichment

Stage 3: AI Personalization

Stage 4: Sequencing

Stage 5: Sending Infrastructure

Stage 6: AI-Powered Follow-Up

The 3-Line Cold Email Framework

Examples

Troubleshooting

Related Skills

SkillSafety

Source

Other skills for the same job

More in the registry