Ready to supercharge your AI agents?

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

Scorecard Data Source Reference · brycewang-stanford | DecimalAI

Documentationby brycewang-stanford

Scorecard Data Source Reference

College Scorecard — post-enrollment outcomes linking aid records to IRS/Treasury earnings. Earnings, loan repayment, debt via six Portal sub-datasets.

Never made a case worse

SkillScore

SkillScore100/100

Test results 100

Safety

3/3 safety checks passed

Code scan
Intent
Content

What it changes

43%100%of test cases passed +57 pts

Without skill10 of 23

With skill23 of 23

13 fixed10 passed either way0 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 taskWhich reference file in the Scorecard collection details program-level earnings and cumulative debt outcomes broken down by field of study or CIP code?

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 brycewang-stanford-education-data-source-scorecard
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("brycewang-stanford-education-data-source-scorecard")
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 brycewang-stanford-education-data-source-scorecard

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/brycewang-stanford-education-data-source-scorecard/SKILL.md https://app.decimal.ai/s/brycewang-stanford-education-data-source-scorecard/SKILL.md

Copy SKILL.md

Save it in your project, as

.claude/skills/brycewang-stanford-education-data-source-scorecard/SKILL.md

This gets SKILL.md only. The skill's 7 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 brycewang-stanford-education-data-source-scorecard

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("brycewang-stanford-education-data-source-scorecard")
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/brycewang-stanford-education-data-source-scorecard/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/brycewang-stanford-education-data-source-scorecard/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 7 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 brycewang-stanford-education-data-source-scorecard

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("brycewang-stanford-education-data-source-scorecard")
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/brycewang-stanford-education-data-source-scorecard/SKILL.md

Give it to your agent

from pathlib import Path
from agents import Agent, Runner

skill = Path(".agents/skills/brycewang-stanford-education-data-source-scorecard/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 7 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 brycewang-stanford-education-data-source-scorecard

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("brycewang-stanford-education-data-source-scorecard")
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/brycewang-stanford-education-data-source-scorecard/SKILL.md

Give it to your agent

from pathlib import Path
from openai import OpenAI

client = OpenAI()
skill = Path(".agents/skills/brycewang-stanford-education-data-source-scorecard/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 7 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 brycewang-stanford-education-data-source-scorecard

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("brycewang-stanford-education-data-source-scorecard")
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/brycewang-stanford-education-data-source-scorecard/SKILL.md

Give it to your agent

from pathlib import Path
import anthropic

client = anthropic.Anthropic()
skill = Path(".agents/skills/brycewang-stanford-education-data-source-scorecard/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 7 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 brycewang-stanford-education-data-source-scorecard

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("brycewang-stanford-education-data-source-scorecard")
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/brycewang-stanford-education-data-source-scorecard/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/brycewang-stanford-education-data-source-scorecard/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 7 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 brycewang-stanford-education-data-source-scorecard

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("brycewang-stanford-education-data-source-scorecard")
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/brycewang-stanford-education-data-source-scorecard/SKILL.md

Give it to your agent

from pathlib import Path
from langchain.agents import create_agent

skill = Path(".agents/skills/brycewang-stanford-education-data-source-scorecard/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 7 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 brycewang-stanford-education-data-source-scorecard

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("brycewang-stanford-education-data-source-scorecard")
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/brycewang-stanford-education-data-source-scorecard/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 7 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 brycewang-stanford-education-data-source-scorecard

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("brycewang-stanford-education-data-source-scorecard")
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/brycewang-stanford-education-data-source-scorecard/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/brycewang-stanford-education-data-source-scorecard/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 7 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 brycewang-stanford-education-data-source-scorecard

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("brycewang-stanford-education-data-source-scorecard")
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/brycewang-stanford-education-data-source-scorecard/SKILL.md

Give it to your agent

from pathlib import Path
from pydantic_ai import Agent

skill = Path(".agents/skills/brycewang-stanford-education-data-source-scorecard/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 7 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 brycewang-stanford-education-data-source-scorecard

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("brycewang-stanford-education-data-source-scorecard")
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/brycewang-stanford-education-data-source-scorecard/SKILL.md

Give it to your agent

from pathlib import Path
from deepagents import create_deep_agent

skill = Path(".agents/skills/brycewang-stanford-education-data-source-scorecard/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 7 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 brycewang-stanford-education-data-source-scorecard

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("brycewang-stanford-education-data-source-scorecard")
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/brycewang-stanford-education-data-source-scorecard/SKILL.md

Give it to your agent

import asyncio
from pathlib import Path
from agent_framework import Agent

skill = Path(".agents/skills/brycewang-stanford-education-data-source-scorecard/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 7 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 brycewang-stanford-education-data-source-scorecard

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("brycewang-stanford-education-data-source-scorecard")
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/brycewang-stanford-education-data-source-scorecard/SKILL.md

Give it to your agent

from pathlib import Path
from strands import Agent

skill = Path(".agents/skills/brycewang-stanford-education-data-source-scorecard/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 7 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 brycewang-stanford-education-data-source-scorecard

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/brycewang-stanford-education-data-source-scorecard/SKILL.md https://app.decimal.ai/s/brycewang-stanford-education-data-source-scorecard/SKILL.md

Copy SKILL.md

Save it in your project, as

.agents/skills/brycewang-stanford-education-data-source-scorecard/SKILL.md

This gets SKILL.md only. The skill's 7 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/brycewang-stanford-education-data-source-scorecard/SKILL.md

Read the summary as JSON

curl -fsSL https://app.decimal.ai/s/brycewang-stanford-education-data-source-scorecard.json

This gets SKILL.md only. The skill's 7 other files are not included.

Other ways to add it

Other setups docs

You don't need a DecimalAI account.

Version: v1, tested 8/9/2026
Source: brycewang-stanford/Auto-Empirical-Research-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

Scorecard Data Source Reference

College Scorecard — the primary institutional-level source for post-enrollment labor market outcomes, linking NSLDS financial aid records to IRS/Treasury earnings data. Use when comparing institutions on post-graduation earnings, loan repayment, or student debt, or when actual tax-record-based earnings are required rather than survey estimates. Covers six sub-datasets accessed via Portal mirrors. Critical limitation: tracks only Title IV federal aid recipients, not all students.

Federal data on post-college outcomes including earnings, debt, and repayment for students who received Title IV financial aid. Links education records to IRS tax data for actual earnings, making it the primary source for post-college labor market outcomes.

CRITICAL: Value Encoding and Missing Data

The Education Data Portal uses integer encodings for all categorical variables and lowercase, restructured variable names that differ from the original Scorecard column names. Suppression encoding differs by dataset:

Earnings/counts: -3 integer code is the primary suppression indicator

Yes/No flags (institutional characteristics): null for missing, 0/1 for valid

Rates (repayment, default): null for missing

The original Scorecard string "PrivacySuppressed" does NOT appear in Portal data

Context pred_degree_awarded_ipeds HBCU / tribal flags religious_affiliation
Portal (integer) 0-4 0 / 1 Integer codes 22-200
Original Scorecard String labels String labels String labels

See ./references/variable-definitions.md for complete encoding tables.

Context	`pred_degree_awarded_ipeds`	HBCU / tribal flags	`religious_affiliation`
Portal (integer)	`0`-`4`	`0` / `1`	Integer codes 22-200
Original Scorecard	String labels	String labels	String labels

What is College Scorecard?

Publisher: U.S. Department of Education
Primary value: Post-college labor market outcomes (earnings) and debt/repayment metrics
Data sources: NSLDS (loans/aid), IRS/Treasury (earnings), IPEDS (institutional characteristics)
Coverage: Title IV federal aid recipients only — not all students
Unique feature: Links education to IRS tax records for actual earnings data
Access: Education Data Portal mirrors (parquet/CSV); see datasets-reference.md for paths, mirrors.yaml for mirror config, fetch-patterns.md for fetch code
Primary identifier: unitid (IPEDS institution ID)

Reference File Structure

File	Purpose	When to Read
`earnings-data.md`	Post-college earnings methodology, cohorts, time horizons	Analyzing earnings outcomes
`debt-repayment.md`	Student debt, repayment rates, default rates	Analyzing debt or loan outcomes
`completion-rates.md`	Completion metrics vs IPEDS	Comparing graduation rates
`population-coverage.md`	Title IV limitation details, who is included/excluded	Understanding data representativeness
`variable-definitions.md`	Key variables, naming conventions, special values	Building queries or interpreting results
`data-quality.md`	Suppression rules, selection bias, known limitations	Assessing data reliability
`field-of-study.md`	Program-level earnings and debt data	Analyzing outcomes by major/CIP code

Decision Trees

What outcome am I researching?

Outcome type?
├─ Post-college earnings
│   ├─ Institution-level → ./references/earnings-data.md
│   └─ By field of study → ./references/field-of-study.md
├─ Student debt levels
│   ├─ Cumulative borrowing → ./references/debt-repayment.md
│   └─ Debt by field → ./references/field-of-study.md
├─ Loan repayment/default
│   └─ Repayment rates → ./references/debt-repayment.md
├─ Completion rates
│   └─ Scorecard completion → ./references/completion-rates.md
└─ Understanding limitations
    ├─ Who is included → ./references/population-coverage.md
    └─ Data quality issues → ./references/data-quality.md

How do I interpret this data?

Interpretation question?
├─ Why are earnings suppressed?
│   └─ Privacy thresholds → ./references/data-quality.md
├─ What does "6-year earnings" mean?
│   └─ Cohort timing → ./references/earnings-data.md
├─ Why don't Scorecard rates match IPEDS?
│   └─ Different cohorts → ./references/completion-rates.md
├─ What loans are included in debt?
│   └─ Federal only → ./references/debt-repayment.md
└─ How representative is this data?
    └─ Title IV coverage → ./references/population-coverage.md

Building a query?

Query construction?
├─ Variable names and codes → ./references/variable-definitions.md
├─ Suppression flags to handle → ./references/data-quality.md
├─ Understanding cohort years → ./references/earnings-data.md
└─ Field-level queries → ./references/field-of-study.md

Quick Reference: Scorecard Variables

Portal Data Structure (CRITICAL)

The Portal uses LONG format with time horizon as a column, NOT the WIDE format from original Scorecard bulk download files. Portal column names are all lowercase and differ significantly from original Scorecard names.

Original Scorecard (WIDE)	Portal Column (LONG)	How to Get
`MD_EARN_WNE_P6`	`earnings_med`	Filter: `years_after_entry == 6`
`MD_EARN_WNE_P10`	`earnings_med`	Filter: `years_after_entry == 10`
`COUNT_WNE_P6`	`count_working`	Filter: `years_after_entry == 6`
`MN_EARN_WNE_P6`	`earnings_mean`	Filter: `years_after_entry == 6`
`CONTROL`, `INSTNM`	NOT IN EARNINGS	Join to IPEDS directory or `inst_characteristics` dataset
`CDR3`	`default_rate`	In `repayment_fsa` dataset; filter: `years_since_entering_repay`
`RPY_3YR_RT`	`repay_rate`	In `repayment_nslds` dataset; filter: `years_since_entering_repay == 3`

Earnings Dataset Columns (Actual Portal Names)

Source dataset: scorecard/colleges_scorecard_earnings (203,066 rows x 33 columns)

Portal Column	Type	Description	Original Scorecard
`unitid`	Int64	IPEDS institution ID	`UNITID`
`opeid`	String	OPE ID (8-digit, zero-padded)	`OPEID`
`year`	Int64	Data year (2003-2014, 2018)	File year
`years_after_entry`	Int64	Years since first enrollment (6-10)	Encoded in variable name
`cohort_year`	Int64	Entry cohort year	Encoded in variable name
`earnings_med`	Int64	Median earnings (W-2)	`MD_EARN_WNE_P*`
`earnings_mean`	Int64	Mean earnings	`MN_EARN_WNE_P*`
`earnings_sd`	Int64	Standard deviation of earnings	`SD_EARN_WNE_P*`
`earnings_pct10`	Int64	10th percentile earnings	`PCT10_EARN_WNE_P*`
`earnings_pct25`	Int64	25th percentile earnings	`PCT25_EARN_WNE_P*`
`earnings_pct75`	Int64	75th percentile earnings	`PCT75_EARN_WNE_P*`
`earnings_pct90`	Int64	90th percentile earnings	`PCT90_EARN_WNE_P*`
`count_working`	Int64	Count working and not enrolled	`COUNT_WNE_P*`
`count_not_working`	Int64	Count not working and not enrolled	`COUNT_NWNE_P*`
`earnings_greater_than_25k_pct`	Float64	Share earning > $25K	`GT_25K_P*`
`earnings_lowinc_mean`	Int64	Mean earnings, low-income	`MN_EARN_WNE_INC1_P*`
`earnings_midinc_mean`	Int64	Mean earnings, mid-income	`MN_EARN_WNE_INC2_P*`
`earnings_highinc_mean`	Int64	Mean earnings, high-income	`MN_EARN_WNE_INC3_P*`
`earnings_dep_mean`	Int64	Mean earnings, dependent students	—
`earnings_dep_lowinc_mean`	Int64	Mean earnings, dependent low-income	—
`earnings_ind_mean`	Int64	Mean earnings, independent students	—
`earnings_female_mean`	Int64	Mean earnings, female	—
`earnings_male_mean`	Int64	Mean earnings, male	—
`count_working_*`	Int64	Count working by subgroup	—

Key Identifiers

ID	Format	Level	Example	Notes
`unitid`	6-digit integer	Institution	`110635`	Same as IPEDS unitid; primary join key
`opeid`	8-digit string	OPE (Title IV)	`"00100200"`	Zero-padded; present in all datasets
`opeid6`	Integer	6-digit OPE	`1002`	Numeric, no zero-padding

Data Timing

Metric	Dimension Column	Values	Typical Lag
Earnings	`years_after_entry`	6, 7, 8, 9, 10	Data from 7+ years ago
Default	`years_since_entering_repay`	2, 3	Varies
Repayment	`years_since_entering_repay`	1, 3, 5, 7	Varies

"After entry" means after first enrollment, not after graduation.

Categorical Value Encodings (Institutional Characteristics Dataset)

Variable	Values
`pred_degree_awarded_ipeds`	0=Not classified, 1=Certificate, 2=Associate's, 3=Bachelor's, 4=Graduate
Yes/No flags (HBCU, tribal, etc.)	0=No, 1=Yes, null=Missing
`religious_affiliation`	76 integer codes 22-200 (see variable-definitions.md for complete mapping), null=None/Missing

Missing Data Codes

Code	Meaning	Which Datasets
`-3`	Suppressed for privacy	Earnings dataset (earnings and count columns) — primary suppression indicator
`null`	Missing/not applicable	Institutional characteristics (yes/no flags), repayment/default (rates)
Positive numeric	Actual value	Earnings, debt, counts, rates

import polars as pl

# Filter for valid earnings (handle -3 suppression code)
valid = df.filter(
    (pl.col("earnings_med").is_not_null()) &
    (pl.col("earnings_med") != -3)
)

# Filter for 6-year earnings specifically
six_yr_valid = valid.filter(pl.col("years_after_entry") == 6)

Data Access

Datasets for Scorecard are available via the mirror system. See datasets-reference.md for canonical paths, mirrors.yaml for mirror configuration, and fetch-patterns.md for fetch code patterns.

Codebooks are .xls files co-located with data in all mirrors. Use get_codebook_url() from fetch-patterns.md to construct download URLs.

Truth Hierarchy: When interpreting variable values, apply this priority:

Actual data file (what you observe in the parquet/CSV) — this IS the truth

Live codebook (.xls in mirror) — authoritative documentation, may lag

This skill documentation — convenient summary, may drift from codebook

If this documentation contradicts the codebook, trust the codebook. If the codebook contradicts observed data, trust the data and investigate.

All Scorecard Datasets (6 total)

Dataset	Path	Codebook	Type	Years
Earnings	`scorecard/colleges_scorecard_earnings`	`scorecard/codebook_colleges_scorecard_earnings`	Single	varies
Default	`scorecard/colleges_scorecard_repayment_fsa`	`scorecard/codebook_colleges_scorecard_default`	Single	1996-2020
Institutional Characteristics	`scorecard/colleges_scorecard_inst_characteristics`	`scorecard/codebook_colleges_scorecard_institutional-characteristics`	Single	1996-2020
Repayment	`scorecard/colleges_scorecard_repayment_nslds`	`scorecard/codebook_colleges_scorecard_repayment`	Single	2007-2016
Student Characteristics (Aid)	`scorecard/colleges_scorecard_student_body_nslds`	`scorecard/codebook_colleges_scorecard_student-characteristics_aid-applicants`	Single	1997-2016
Student Characteristics (Home)	`scorecard/colleges_scorecard_student_body_treasury`	`scorecard/codebook_colleges_scorecard_student-characteristics_home-neighborhood`	Single	1997-2016

Scorecard naming note: Data file paths differ significantly from codebook paths. Notable mismatches: data repayment_fsa vs codebook default; data inst_characteristics vs codebook institutional-characteristics; data repayment_nslds vs codebook repayment; data student_body_nslds vs codebook student-characteristics_aid-applicants; data student_body_treasury vs codebook student-characteristics_home-neighborhood. Always use the exact paths shown above.

Fetching Data

import polars as pl
from fetch_utils import fetch_from_mirrors  # See fetch-patterns.md

# Fetch earnings data
earnings = fetch_from_mirrors("scorecard/colleges_scorecard_earnings")

# Filter by time horizon (LONG format — filter, don't use wide column names)
six_yr = earnings.filter(pl.col("years_after_entry") == 6)

# Filter for valid earnings (exclude -3 suppression code)
valid = six_yr.filter(
    (pl.col("earnings_med").is_not_null()) &
    (pl.col("earnings_med") != -3)
)

# Institution names/control are NOT in the earnings dataset.
# Join to inst_characteristics or IPEDS directory:
inst = fetch_from_mirrors("scorecard/colleges_scorecard_inst_characteristics",
                          years=[2020])
valid = valid.join(
    inst.select("unitid", "inst_name", "pred_degree_awarded_ipeds"),
    on="unitid", how="left"
)

Common Pitfalls

Pitfall	Issue	Solution
"All graduates" claims	Scorecard covers Title IV recipients only, not all students	Note Title IV limitation prominently in any analysis
Wage comparison	Comparing to BLS wages or Census income uses different populations	Use for relative comparisons, not absolute claims; document population differences
Ignoring suppression	Many programs have no data due to privacy thresholds	Check suppression rates before analyzing; document coverage
Time lag ignored	Earnings reflect old cohorts (6-year = data from 7+ years ago)	Document data vintage and cohort years explicitly
Total borrowing assumption	Scorecard debt includes only federal loans, not private	State "federal loans only" when reporting debt figures
String codes from docs	Original Scorecard uses string labels; Portal uses integers	Verify actual data types in Portal parquet files; use integer codes
Wide-format variable names	Using `MD_EARN_WNE_P10` column name on Portal data	Portal uses LONG format — filter `years_after_entry` instead
Assuming null = suppressed	Earnings dataset uses `-3` for suppression, not null	Filter both: `is_not_null() & != -3`
Using uppercase names	Original Scorecard uses `MD_EARN_WNE_P6`; Portal uses `earnings_med`	Always use lowercase Portal names from actual data

Critical Limitation: Title IV Recipients Only

The single most important caveat for all Scorecard analysis:

Scorecard tracks ONLY students who received federal financial aid (Title IV):

Pell Grants
Federal student loans (Direct, Perkins, PLUS)
Federal work-study

Excluded Group	Impact
Full-pay students	Often higher-income; different outcomes
Students with only state/institutional aid	Missing from data
International students	Not eligible for federal aid
Some graduate students	If they received no federal aid

Coverage varies dramatically by institution type:

Institution Type	Typical Title IV Coverage
For-profit colleges	80-90%+
Community colleges	60-80%
Public flagships	50-70%
Selective private colleges	30-50%

Data systematically overrepresents lower-income students who are more likely to need federal aid.

What Scorecard Data Does NOT Include

Excluded	Why It Matters
Non-Title IV students	Often higher-income; different outcomes
Self-employment income	1099 income excluded from earnings
Students still in school	Not working = not in earnings data
Private student loans	Only federal loans tracked
Students who left the country	Lost to follow-up

Comparison: Scorecard vs IPEDS

Aspect	College Scorecard	IPEDS
Who's tracked	Title IV aid recipients	First-time, full-time students
Includes part-time	Yes	No (for grad rates)
Includes transfers-in	Yes	No (tracked at origin)
Outcome focus	Earnings, debt, repayment	Completion, retention
Data source	NSLDS + IRS	Institution-reported

Related Data Sources

Source	Relationship	When to Use
`education-data-source-ipeds`	Institutional characteristics, enrollment, finance	Join on `unitid` for institution names, control type, enrollment context
`education-data-source-pseo`	Alternative post-college earnings (Census LEHD)	When broader population coverage needed (not limited to Title IV)
`education-data-source-fsa`	Federal student aid details	Deeper analysis of aid types and disbursements
`education-data-explorer`	Parent discovery skill	Finding available endpoints
`education-data-query`	Data fetching	Downloading parquet/CSV files

Topic Index

Topic	Reference File
Earnings methodology	`./references/earnings-data.md`
Cohort definitions	`./references/earnings-data.md`
IRS data matching	`./references/earnings-data.md`
Earnings suppression	`./references/data-quality.md`
Debt metrics	`./references/debt-repayment.md`
Repayment rates	`./references/debt-repayment.md`
Default rates	`./references/debt-repayment.md`
NSLDS data	`./references/debt-repayment.md`
Completion methodology	`./references/completion-rates.md`
IPEDS comparison	`./references/completion-rates.md`
Title IV coverage	`./references/population-coverage.md`
Who is excluded	`./references/population-coverage.md`
Selection bias	`./references/population-coverage.md`
Variable names	`./references/variable-definitions.md`
Special values	`./references/variable-definitions.md`
Privacy suppression	`./references/data-quality.md`
Data limitations	`./references/data-quality.md`
Program-level data	`./references/field-of-study.md`
CIP codes	`./references/field-of-study.md`

When it activates

Automatically, when a task matches. Use when tax-record-based earnings needed. Tracks only Title IV aid recipients, not all students.

About this skill

Automatic

Data Analyst pack

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

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

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

brycewang-stanford / Auto-Empirical-Research-Skills

skills/17-DAAF-Contribution-Community-daaf/dot-claude/skills/education-data-source-scorecard/SKILL.md

View on GitHub ↗

License NOASSERTIONImported 8/2/2026Version v1Bundled 7 files

Illustrative Want to see what brycewang-stanford/education-data-source-scorecard would put in front of your model? Preview the rendered body and check its trigger phrases — no LLM call, no signup.

Related skills

Skills that do similar work, with their headline benchmark lift.

From the same source and nearby in the catalogue.

Ready to supercharge your AI agents?

SkillScore

Safety

What it changes

Track record

Use this skill

Scorecard Data Source Reference

What is College Scorecard?

Reference File Structure

Decision Trees

What outcome am I researching?

How do I interpret this data?

Building a query?

Quick Reference: Scorecard Variables

Portal Data Structure (CRITICAL)

Earnings Dataset Columns (Actual Portal Names)

Key Identifiers

Data Timing

Categorical Value Encodings (Institutional Characteristics Dataset)

Missing Data Codes

Data Access

All Scorecard Datasets (6 total)

Fetching Data

Common Pitfalls

Critical Limitation: Title IV Recipients Only

What Scorecard Data Does NOT Include

Comparison: Scorecard vs IPEDS

Related Data Sources

Topic Index

When it activates

SkillSafety

Source

Related skills

More in the registry