Ready to supercharge your AI agents?

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

Outreach Manager · aaron-he-zhu | DecimalAI

Marketingby aaron-he-zhu

Outreach Manager

Use when the user asks to "write influencer outreach", "follow up with a creator", "pitch a journalist, hunter, or launch partner", or "negotiate partnership terms"; produces personalized pitches, multi-touch follow-up sequences, negotiation scripts with objection handling, and a status pipeline tracker — the shared outreach mechanics engine for creator, media/analyst, launch-partner, and social-selling / advocate-recruitment targets. Not for finalizing signed agreements — use contract-helper; n

Never made a case worseMaintained

SkillScore

SkillScore95/100

Test results 95

Safety

3/3 safety checks passed

Code scan
Intent
Content

What it changes

41%95%of test cases passed +55 pts

Without skill9 of 22

With skill21 of 22

12 fixed9 passed either way1 still failing0 broke

Measured on gemini-3.6-flash9/9/2026v2

Track record

Agent runs: 0; All time

See one task, with and without the skill

One test taskThrough repeated campaigns, we identified top objection patterns and confirmed response rate baselines for our SaaS productivity tool outreach. We want to promote these durable performance insights to memory/hot-cache.md. What gating condition applies to this memory promotion?

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 aaron-he-zhu-outreach-manager
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("aaron-he-zhu-outreach-manager")
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 aaron-he-zhu-outreach-manager

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/aaron-he-zhu-outreach-manager/SKILL.md https://app.decimal.ai/s/aaron-he-zhu-outreach-manager/SKILL.md

Copy SKILL.md

Save it in your project, as

.claude/skills/aaron-he-zhu-outreach-manager/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 aaron-he-zhu-outreach-manager

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("aaron-he-zhu-outreach-manager")
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/aaron-he-zhu-outreach-manager/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/aaron-he-zhu-outreach-manager/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 aaron-he-zhu-outreach-manager

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("aaron-he-zhu-outreach-manager")
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/aaron-he-zhu-outreach-manager/SKILL.md

Give it to your agent

from pathlib import Path
from agents import Agent, Runner

skill = Path(".agents/skills/aaron-he-zhu-outreach-manager/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 aaron-he-zhu-outreach-manager

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("aaron-he-zhu-outreach-manager")
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/aaron-he-zhu-outreach-manager/SKILL.md

Give it to your agent

from pathlib import Path
from openai import OpenAI

client = OpenAI()
skill = Path(".agents/skills/aaron-he-zhu-outreach-manager/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 aaron-he-zhu-outreach-manager

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("aaron-he-zhu-outreach-manager")
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/aaron-he-zhu-outreach-manager/SKILL.md

Give it to your agent

from pathlib import Path
import anthropic

client = anthropic.Anthropic()
skill = Path(".agents/skills/aaron-he-zhu-outreach-manager/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 aaron-he-zhu-outreach-manager

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("aaron-he-zhu-outreach-manager")
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/aaron-he-zhu-outreach-manager/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/aaron-he-zhu-outreach-manager/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 aaron-he-zhu-outreach-manager

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("aaron-he-zhu-outreach-manager")
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/aaron-he-zhu-outreach-manager/SKILL.md

Give it to your agent

from pathlib import Path
from langchain.agents import create_agent

skill = Path(".agents/skills/aaron-he-zhu-outreach-manager/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 aaron-he-zhu-outreach-manager

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("aaron-he-zhu-outreach-manager")
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/aaron-he-zhu-outreach-manager/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 aaron-he-zhu-outreach-manager

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("aaron-he-zhu-outreach-manager")
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/aaron-he-zhu-outreach-manager/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/aaron-he-zhu-outreach-manager/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 aaron-he-zhu-outreach-manager

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("aaron-he-zhu-outreach-manager")
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/aaron-he-zhu-outreach-manager/SKILL.md

Give it to your agent

from pathlib import Path
from pydantic_ai import Agent

skill = Path(".agents/skills/aaron-he-zhu-outreach-manager/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 aaron-he-zhu-outreach-manager

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("aaron-he-zhu-outreach-manager")
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/aaron-he-zhu-outreach-manager/SKILL.md

Give it to your agent

from pathlib import Path
from deepagents import create_deep_agent

skill = Path(".agents/skills/aaron-he-zhu-outreach-manager/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 aaron-he-zhu-outreach-manager

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("aaron-he-zhu-outreach-manager")
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/aaron-he-zhu-outreach-manager/SKILL.md

Give it to your agent

import asyncio
from pathlib import Path
from agent_framework import Agent

skill = Path(".agents/skills/aaron-he-zhu-outreach-manager/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 aaron-he-zhu-outreach-manager

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("aaron-he-zhu-outreach-manager")
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/aaron-he-zhu-outreach-manager/SKILL.md

Give it to your agent

from pathlib import Path
from strands import Agent

skill = Path(".agents/skills/aaron-he-zhu-outreach-manager/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 aaron-he-zhu-outreach-manager

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/aaron-he-zhu-outreach-manager/SKILL.md https://app.decimal.ai/s/aaron-he-zhu-outreach-manager/SKILL.md

Copy SKILL.md

Save it in your project, as

.agents/skills/aaron-he-zhu-outreach-manager/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/aaron-he-zhu-outreach-manager/SKILL.md

Read the summary as JSON

curl -fsSL https://app.decimal.ai/s/aaron-he-zhu-outreach-manager.json

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

Other ways to add it

Other setups docs

You don't need a DecimalAI account.

Version: v2, tested 9/9/2026
Source: aaron-he-zhu/aaron-marketing-skills ↗
License: Apache-2.0

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

Outreach Manager

Craft personal, professional, persistent outreach; manage negotiations; track relationship progress. Home discipline is influencer (creator outreach), and the same mechanics engine — personalization, multi-touch cadence, negotiation scripts, pipeline tracking — serves media/analyst and launch-partner targets when press-media-relations hands over its media list, angles, and embargo terms, and social-selling / advocate-recruitment targets when social-selling-planner or advocacy-program-designer hands over a warm 1:1. The list, the angle, and the embargo stay with the handing-off skill; this skill owns the pitch execution.

Quick Start

Shortest invocation:

Write an outreach message to @[influencer] for [campaign]

Negotiate a gap between ask and budget:

Help me negotiate with @[influencer] who is asking for $[X] when our budget is $[Y]

Skill Contract

Reads: a stable opaque creator_ref when one is carried by an authorized upstream artifact, plus transient target handle/profile locator(s), platform, follower count, niche; campaign and product context; compensation type and budget; deliverables and timeline; jurisdiction, intended channel, lawful-basis reference, contact-source/eligibility evidence, and pseudonymous consent subject ID; any prior contact history supplied by the user or loaded from memory. For a rostered creator, reuse a creator-registry aggregate ID only when its identity link is verified, resolve it through the authorized artifact or verified registry link, then check memory/creators/<aggregate-id>.md — the creator-registry projection — for the confirmed contact path, last agreed rate, and negotiation/response history. Never derive the path from a raw handle.
Writes: return the outreach artifact inline by default; save it to memory/influencer/outreach-manager/YYYY-MM-DD-<topic>.md only with exact WARM-save authorization. Reuse an explicitly carried opaque creator_ref, or a verified creator-registry aggregate ID; otherwise generate one random creator-<UUIDv4> once for this lineage. Never derive creator_ref from a raw handle, name, profile URL, email, provider ID, or deterministic hash. Before saving, replace raw locators and resolved recipient identities with the stable creator_ref and resolvable opaque recipient_ref, contact_source_ref, handle_ref, and evidence/approval references. A raw URL or handle is never a persisted source_ref. When no authorized source artifact or verified registry link can resolve an opaque identity/contact ref, set identity_status: unresolved, save no hidden raw-locator mapping, set cross_session_locator_required: true, and request the transient locator again at dispatch. Never persist a raw handle, name, profile URL, email, phone, postal address, provider ID, or credential. When a cycle closes, each outcome update (final agreed rate, response history, confirmed contact-path reference) requires a separate exact authorization for an operation: propose request to memory/events/creators.ndjson; only creator-registry writes canonical roster records.
Promotes: only with separate exact authorization, promote durable facts (confirmed partners, agreed rates, top objection patterns, response-rate baselines) to memory/hot-cache.md.
Done when:
- A personalized pitch (plus at least one variation) exists for each target influencer.
- Every personalization fact records source_ref and observed_at; unverifiable facts remain placeholders or are omitted.
- A follow-up cadence and pipeline status are recorded for every contacted creator, including terminal no-contact states.
- Jurisdiction, channel, lawful basis, and contact eligibility are explicit; any send/schedule remains gated by exact approval and a live suppression check.
- Confirmed partners are flagged with agreed terms for handoff.
Primary next skill: contract-helper

Handoff Summary

Emit the standard shape from skill-contract.md §Handoff Summary Format.

Data Sources

Drafting needs no live integration (Tier 1). The skill works from inputs you provide — paste the influencer handles, follower counts, niche, budget, and deliverables, and it can produce a clearly labeled draft without any tool connection. Sending or scheduling is different: it requires the pseudonymous subject ID and a live consent-registry suppression query immediately before dispatch; an unavailable or failed query is not permission to send.

Where a connector could speed up the work, use these ~~ placeholders:

~~influencer database — pull handle, follower count, niche, and past partnerships instead of typing them.
~~social platform analytics — verify audience demographics and recent posts for personalization.
~~CRM — sync pipeline status, last-contact dates, and next actions.
~~email/DM tool — dispatch only the single currently due, exactly approved touch after its fresh eligibility and live-suppression checks. Future touches remain unscheduled drafts regardless of provider features.

See CONNECTORS.md for the free/keyless recipe per category. No integration is required; when one is absent, ask the user for the inputs directly. Missing target or campaign details do not block a reversible draft: use explicit bracketed placeholders, generalize any unverified personalization, and identify the smallest inputs needed to finish it.

Instructions

Runtime Reads

../../../references/runtime-invocation.md
../../../references/registry-event-protocol.md

Procedure

When a user requests outreach help, run these steps. Each step has a fill-in template in references/templates.md — copy the matching block and replace the placeholders. Apply the hard copy rules in references/cold-copy-rules.md before any message ships.

Draft/send boundary: drafting is reversible; sending or scheduling is an external side effect. Even when the shortlist or personalization facts are missing, produce an inline first-touch draft labeled DRAFT — NOT SENT with explicit placeholders and no invented facts, then state what is still needed for personalization. Never stop with only an input request when a safe placeholder draft can be produced.

Do not send or schedule anything until all of these are true:

The exact recipient (not a segment or placeholder), intended channel, final message, and—when scheduling that one touch—one concrete ISO-8601 dispatch_at timestamp plus timezone are shown and explicitly approved. Approval covers only the single currently due touch. Draft windows such as Day 3–4 are not exact approval, and no approval or provider feature authorizes pre-scheduling later cadence touches.
Jurisdiction, channel, lawful basis with its evidence reference, and contact eligibility are recorded. Unknown or missing eligibility fails closed; this skill does not invent a legal basis or provide legal advice.
Inside each actual delivery job, after resolving the transient recipient and immediately before the provider send call, resolve and verify the bundle runtime as shown below, then run live consent-registry is-suppressed. A suppressed result, missing subject ID, inaccessible/corrupt history, runtime/schema failure, or query error stops that dispatch. A not suppressed result removes only the suppression block; it does not itself prove lawful basis or authorize contact. Never reuse a cached result across cadence touches.

AARON_SKILLS_ROOT="${CLAUDE_PLUGIN_ROOT:-$(git rev-parse --show-toplevel 2>/dev/null || true)}"
PROJECT_ROOT="${PROJECT_ROOT:-}"
SUBJECT_ID="${SUBJECT_ID:-}"
case "$PROJECT_ROOT" in
  /*) ;;
  *) echo "PROJECT_ROOT must be an absolute project path; stop dispatch." >&2; exit 1 ;;
esac
if [ -z "$AARON_SKILLS_ROOT" ] || [ -z "$PROJECT_ROOT" ] || [ -z "$SUBJECT_ID" ] || \
   [ ! -d "$PROJECT_ROOT" ] || [ ! -f "$AARON_SKILLS_ROOT/.claude-plugin/plugin.json" ] || \
   [ ! -f "$AARON_SKILLS_ROOT/references/system-catalog.json" ] || \
   [ ! -f "$AARON_SKILLS_ROOT/references/capability-profiles.json" ] || \
   [ ! -f "$AARON_SKILLS_ROOT/references/registry-event.schema.json" ] || \
   [ ! -f "$AARON_SKILLS_ROOT/scripts/profile-resolver.py" ] || \
   [ ! -f "$AARON_SKILLS_ROOT/scripts/registry-events.py" ]; then
  echo "Verified Aaron Marketing Skills consent runtime unavailable; stop dispatch." >&2
  exit 1
fi
python3 "$AARON_SKILLS_ROOT/scripts/profile-resolver.py" \
  --root "$PROJECT_ROOT" --bundle-root "$AARON_SKILLS_ROOT" diagnose --json >/dev/null || exit 1
SUPPRESSION_JSON="$(python3 "$AARON_SKILLS_ROOT/scripts/registry-events.py" \
  --root "$PROJECT_ROOT" is-suppressed "$SUBJECT_ID")" || exit 1
printf '%s\n' "$SUPPRESSION_JSON" | python3 -c \
  'import json, sys; value = json.load(sys.stdin); raise SystemExit(0 if value.get("aggregate_id") == sys.argv[1] and value.get("suppressed") is False else 1)' \
  "$SUBJECT_ID" || { echo "Recipient is suppressed or suppression result is invalid; stop dispatch." >&2; exit 1; }

PROJECT_ROOT must be supplied by the host as the verified absolute user-project root; do not infer it from the skill bundle, $PWD, a raw event, or user-controlled outreach content. When the subject ID, project root, runtime, or another send-gate input is missing, return a NEEDS_INPUT consent-check handoff to consent-registry containing only the pseudonymous subject ID if supplied, required project-root/runtime capability, intended channel, eligibility evidence refs, and the pending dispatch ref; do not claim that is-suppressed ran or automatically resume delivery. Always leave later cadence touches as unscheduled drafts. When the next touch becomes due, resolve its transient recipient again, obtain fresh exact approval for that touch, repeat every eligibility/runtime/suppression check inside its delivery job, and dispatch only that one touch.

If a recipient declines the offer or says they do not take sponsored work, void every remaining draft touch for that cadence, do not relabel the offer or switch channels, and record only the exact scoped preference inline (campaign/offer/category scope, observed_at, and opaque source_ref); persisting it or proposing it to creator-registry needs separate exact authorization. A prior preference blocks a new cadence inside its recorded scope until newer cited evidence shows the creator reopened that scope; if the old preference's scope or superseding evidence is unknown, fail closed with NEEDS_INPUT rather than inventing a cooling period. Do not call global suppress for a scoped decline or commercial objection alone. An explicit stop-contact request, unsubscribe, verified channel/provider spam complaint, or consent withdrawal voids every remaining draft touch across channels and uses the consent-registry's direct deny-only suppress path with the exact subject-free reason code: stop-contact → user-request; unsubscribe → unsubscribe; verified spam/provider complaint → complaint; consent withdrawal → withdrawal. If the verified runtime is unavailable, emit its exact immediate-suppress-handoff and state that suppression has not yet been recorded.

Gather outreach context and lock identity — capture campaign/product context, transient target handle/profile locator(s), platform, followers, niche, compensation type, budget, deliverables, and timeline. Reuse an explicitly carried opaque creator_ref, or a creator-registry aggregate ID only when its identity link is verified; otherwise generate one random creator-<UUIDv4> once for this lineage. Never use or hash a raw locator into creator_ref. Save an opaque handle_ref, recipient_ref, contact_source_ref, or evidence source_ref only when an authorized source artifact or verified registry link resolves it. Otherwise set identity_status: unresolved, persist no raw locator or hidden mapping, set cross_session_locator_required: true, and require the raw locator again at dispatch. Any saved artifact or handoff contains only opaque identity/contact/evidence refs plus jurisdiction, channel, lawful-basis evidence, and contact-eligibility result—never a raw handle, name, profile/content URL, email, phone, postal address, provider ID, or credential. Re-engaging a verified rostered creator starts from the confirmed contact-path reference and last agreed rate, not a cold pitch. Template: Step 1.
Create personalized outreach — list personalization points (recent content, style, audience, values, past partners), and attach source_ref plus observed_at to every factual point before using it. Omit or generalize anything unverifiable; never invent personal viewing, purchasing, product-use, or relationship history. Then write the primary message plus a DM-friendly short version and a formal email/management version. Template: Step 2. Media/analyst/hunter targets: personalize on beat and recent coverage, lead with the story angle (not a compensation offer), carry the embargo terms verbatim from the press-media-relations artifact, and never invent quotes or data — claims come from the approved message house.
Create follow-up sequence — draft a 4-touch strategy window (Day 0 / 3-4 / 7-8 / 14, then archive at Day 21), each touch adding new evidence-backed or approved value and getting shorter. These ranges are planning guidance only. Always leave future touches unscheduled; when one becomes due, approve/check/dispatch only that touch with one concrete ISO-8601 dispatch_at plus timezone if scheduled. Cap at 3-4 follow-ups only while there is no blocking scoped preference or other negative signal, make it easy to say no, and use only the approved eligible channel. A new channel needs its own eligibility evidence and exact approval. A campaign/offer decline terminates that cadence and records the scoped preference inline; persisting it needs separate exact authorization. A later cadence within that scope requires newer reopening evidence. An explicit stop-contact (user-request), unsubscribe (unsubscribe), verified spam/provider complaint (complaint), or consent withdrawal (withdrawal) additionally triggers the suppression path above. Template: Step 3.
Provide negotiation support — map the ask/budget gap, then apply value-exchange, scope-adjustment, or future-value strategies with ready scripts and an objection/response table. Template: Step 4.
Track outreach pipeline — record stage counts and conversion rates, a per-creator detailed pipeline, today's prioritized actions, and pipeline health (response rate, confirmation rate, time-to-confirm, top objection). Record offer declines as terminal for that cadence with a scoped preference. Record explicit stop-contact (user-request), unsubscribe (unsubscribe), verified channel/provider spam complaint (complaint), or consent withdrawal (withdrawal) as terminal no-contact states with the suppression event/handoff reference. Never leave a follow-up action on either row. Template: Step 5. Active-cycle tracking lives here; when a cycle closes (confirmed or archived), offer the closed outcome as a one-line operation: propose update for separate exact authorization so creator-registry can reconcile it.

Example

User: "Write outreach for creator_ref: creator-042. I supplied a transient content locator plus source_ref: [opaque authorized evidence ref], observed_at: [ISO 8601], and observable_detail: [exact visible detail]; use only the approved campaign, offer, and product wording in [claims-or-brief-ref]."

Output (abridged):

## Outreach for `creator_ref: creator-042`

### Personalization Points
- Content item: `[content title or format supported by content_ref]`
- Observable detail: `[exact visible detail from supplied evidence]`
- Evidence: `source_ref: [opaque authorized evidence ref]`  •  `observed_at: [ISO 8601]`

### Primary Message
Subject: `[evidence-backed content topic]` — collaboration idea from `[Brand]`

`[Recipient name resolved transiently at dispatch]`, your `[content item]` `[observable detail supported by content_ref]`. I'm `[Sender]` from `[Brand]`. `[Approved campaign/product sentence from claims-or-brief-ref]` We're offering `[approved compensation]` for `[approved deliverable]`. Open to hearing more?

Full multi-version output, follow-up cadence, negotiation guide, and pipeline tracker live in references/templates.md.

Reference Materials

references/templates.md — fill-in templates for all five steps, the full worked example, and outreach tips.
references/cold-copy-rules.md — hard cold-outreach copy rules: first-line bans, per-step sentence caps, soft CTAs, observation framing, no link in step 1.
skill-contract.md — shared contract and Handoff Summary format.
state-model.md — memory tiers and save-path conventions.
runtime-invocation.md — safe bundle-root resolution and feature-runtime verification before live suppression queries.
CONNECTORS.md — free/keyless data recipe per connector category.
STAR benchmark scoring at references/star-benchmark.md — quality scoring reference for downstream review.
expert-panel.md — multi-persona review method for pressure-testing outreach copy before sending.
Sibling skills: influencer-discovery, fit-scorer, brief-generator, contract-helper, creator-content-auditor.

Next Best Skill

Primary: contract-helper — once a partner is confirmed, turn the agreed commercial terms and rights into the agreement before content production.
Alternate: creator-content-auditor — only after the agreement/brief exists and the creator has produced draft content, review that content before it ships.
Alternate: brief-generator — send a full campaign brief to a creator who asked for more detail.

Termination note: keep a visited-set. If a skill in this chain was already invoked this session, stop and report chain-complete rather than re-running it. Max handoff depth is 3.

Related Skills

influencer-discovery - Find influencers to reach out to
fit-scorer - Prioritize who to contact first
brief-generator - Send briefs to confirmed partners
contract-helper - Finalize agreements

About this skill

Automatic

Marketing & Content pack

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v5 — 9/9/2026

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

aaron-he-zhu / aaron-marketing-skills

influencer/activate/outreach-manager/SKILL.md

View on GitHub ↗

License Apache-2.0Imported 10/31/2026Version v2Bundled 2 files

Illustrative Want to see what aaron-he-zhu/outreach-manager 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.

aiskillstore/outreach-managerUse when the user asks to "write influencer outreach", "follow up with a creator", "pitch a journalist, hunter, or launch partner", or "negotiate partnership terms"; produces personalized pitches, multi-touch follow-up sequences, negotiation scripts with objection handling, and a status pipeline tracker — the shared outreach mechanics engine for creator, media/analyst, launch-partner, and social-selling / advocate-recruitment targets. Not for finalizing signed agreements — use contract-helper; nNear-identical description+55 pts pass-rate lift

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

Outreach Manager

Quick Start

Skill Contract

Handoff Summary

Data Sources

Instructions

Runtime Reads

Procedure

Example

Reference Materials

Next Best Skill

Related Skills

SkillSafety

Source

Other skills for the same job

Related skills

More in the registry