Ready to supercharge your AI agents?

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

Databricks Cost Leak Hunter · jeremylongshore | DecimalAI

Financeby jeremylongshore

Databricks Cost Leak Hunter

Hunt down Databricks cost leaks — wasted DBUs, idle clusters, oversized SQL warehouses, and untagged runaway spend — and produce a FinOps cost report.

Maintained

SkillScore

SkillScore68/100

Test results 68

Safety

3/3 safety checks passed

Code scan
Intent
Content

What it changes

50%68%of test cases passed +18 pts

Without skill11 of 22

With skill15 of 22

6 fixed9 passed either way5 still failing2 broke

Measured on gemini-3.6-flash8/14/2026v119 comparable

Track record

Agent runs: 0; All time

See one task, with and without the skill

One test taskOur scan identified $15,000/month in idle clusters, $8,000/month in job tier misconfigurations, $6,000/month in low-CPU overprovisioning, and $4,000/month in Photon premium. Create the executive report headline for a $100,000/month workspace.

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 jeremylongshore-databricks-cost-leak-hunter
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("jeremylongshore-databricks-cost-leak-hunter")
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 jeremylongshore-databricks-cost-leak-hunter

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/jeremylongshore-databricks-cost-leak-hunter/SKILL.md https://app.decimal.ai/s/jeremylongshore-databricks-cost-leak-hunter/SKILL.md

Copy SKILL.md

Save it in your project, as

.claude/skills/jeremylongshore-databricks-cost-leak-hunter/SKILL.md

This gets SKILL.md only. The skill's 5 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 jeremylongshore-databricks-cost-leak-hunter

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("jeremylongshore-databricks-cost-leak-hunter")
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/jeremylongshore-databricks-cost-leak-hunter/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/jeremylongshore-databricks-cost-leak-hunter/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 5 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 jeremylongshore-databricks-cost-leak-hunter

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("jeremylongshore-databricks-cost-leak-hunter")
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/jeremylongshore-databricks-cost-leak-hunter/SKILL.md

Give it to your agent

from pathlib import Path
from agents import Agent, Runner

skill = Path(".agents/skills/jeremylongshore-databricks-cost-leak-hunter/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 5 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 jeremylongshore-databricks-cost-leak-hunter

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("jeremylongshore-databricks-cost-leak-hunter")
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/jeremylongshore-databricks-cost-leak-hunter/SKILL.md

Give it to your agent

from pathlib import Path
from openai import OpenAI

client = OpenAI()
skill = Path(".agents/skills/jeremylongshore-databricks-cost-leak-hunter/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 5 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 jeremylongshore-databricks-cost-leak-hunter

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("jeremylongshore-databricks-cost-leak-hunter")
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/jeremylongshore-databricks-cost-leak-hunter/SKILL.md

Give it to your agent

from pathlib import Path
import anthropic

client = anthropic.Anthropic()
skill = Path(".agents/skills/jeremylongshore-databricks-cost-leak-hunter/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 5 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 jeremylongshore-databricks-cost-leak-hunter

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("jeremylongshore-databricks-cost-leak-hunter")
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/jeremylongshore-databricks-cost-leak-hunter/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/jeremylongshore-databricks-cost-leak-hunter/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 5 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 jeremylongshore-databricks-cost-leak-hunter

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("jeremylongshore-databricks-cost-leak-hunter")
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/jeremylongshore-databricks-cost-leak-hunter/SKILL.md

Give it to your agent

from pathlib import Path
from langchain.agents import create_agent

skill = Path(".agents/skills/jeremylongshore-databricks-cost-leak-hunter/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 5 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 jeremylongshore-databricks-cost-leak-hunter

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("jeremylongshore-databricks-cost-leak-hunter")
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/jeremylongshore-databricks-cost-leak-hunter/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 5 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 jeremylongshore-databricks-cost-leak-hunter

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("jeremylongshore-databricks-cost-leak-hunter")
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/jeremylongshore-databricks-cost-leak-hunter/SKILL.md

Give it to your agent

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

skill = Path(".agents/skills/jeremylongshore-databricks-cost-leak-hunter/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 5 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 jeremylongshore-databricks-cost-leak-hunter

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("jeremylongshore-databricks-cost-leak-hunter")
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/jeremylongshore-databricks-cost-leak-hunter/SKILL.md

Give it to your agent

from pathlib import Path
from pydantic_ai import Agent

skill = Path(".agents/skills/jeremylongshore-databricks-cost-leak-hunter/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 5 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 jeremylongshore-databricks-cost-leak-hunter

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("jeremylongshore-databricks-cost-leak-hunter")
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/jeremylongshore-databricks-cost-leak-hunter/SKILL.md

Give it to your agent

from pathlib import Path
from deepagents import create_deep_agent

skill = Path(".agents/skills/jeremylongshore-databricks-cost-leak-hunter/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 5 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 jeremylongshore-databricks-cost-leak-hunter

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("jeremylongshore-databricks-cost-leak-hunter")
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/jeremylongshore-databricks-cost-leak-hunter/SKILL.md

Give it to your agent

import asyncio
from pathlib import Path
from agent_framework import Agent

skill = Path(".agents/skills/jeremylongshore-databricks-cost-leak-hunter/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 5 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 jeremylongshore-databricks-cost-leak-hunter

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("jeremylongshore-databricks-cost-leak-hunter")
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/jeremylongshore-databricks-cost-leak-hunter/SKILL.md

Give it to your agent

from pathlib import Path
from strands import Agent

skill = Path(".agents/skills/jeremylongshore-databricks-cost-leak-hunter/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 5 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 jeremylongshore-databricks-cost-leak-hunter

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/jeremylongshore-databricks-cost-leak-hunter/SKILL.md https://app.decimal.ai/s/jeremylongshore-databricks-cost-leak-hunter/SKILL.md

Copy SKILL.md

Save it in your project, as

.agents/skills/jeremylongshore-databricks-cost-leak-hunter/SKILL.md

This gets SKILL.md only. The skill's 5 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/jeremylongshore-databricks-cost-leak-hunter/SKILL.md

Read the summary as JSON

curl -fsSL https://app.decimal.ai/s/jeremylongshore-databricks-cost-leak-hunter.json

This gets SKILL.md only. The skill's 5 other files are not included.

Other ways to add it

Other setups docs

You don't need a DecimalAI account.

Version: v1, tested 8/14/2026
Source: jeremylongshore/claude-code-plugins-plus-skills ↗
License: MIT

Manage it in DecimalAI

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

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

Use this skill

SKILL.md

Databricks Cost Leak Hunter

Audits a Databricks workspace for real-dollar cost leaks — idle compute, jobs on the wrong SKU, overprovisioned clusters, and the Photon premium paid without the speedup — then emits a CFO-grokkable, dollar-ranked FinOps remediation report.

Overview

This skill finds where Databricks money is leaking and how much, in dollars, per month. Confirmed-spend figures come from the customer's own system.billing.usage joined to system.billing.list_prices — never an estimate. Two of the four categories (overprovisioning, Photon premium) are explicitly modeled/at-risk amounts, labeled as such so a CFO never confuses them with recoverable spend. The skill surfaces four named leak categories, ranks them by monthly dollar impact, and explains each root cause in FinOps language a CFO can act on without an engineer to translate.

It is architecturally distinct from the v1 databricks-cost-tuning skill: that one AUTHORS policy (creates cluster policies, spot configs). This one DETECTS leaks and reports them, dollarized and ranked. The math is deterministic — bundled scripts in scripts/ do the arithmetic so the agent never eyeballs numbers — and deep domain knowledge lives in references/ loaded only when a leak needs it.

The skill uses two data planes. Dollar figures come from the Databricks CLI Statement Execution API (databricks api post /api/2.0/sql/statements) reading system.* — authenticated by the CLI's own DATABRICKS_HOST+DATABRICKS_TOKEN / databricks auth login. The live config/event evidence that explains why a leak exists (auto-termination setting, node type, autoscale floor, pool min_idle) comes from the custom databricks-workspace-mcp control-plane tools — the one MCP dependency. The SQL produces the number; the workspace MCP turns it into a verified, single-config-change fix.

Prerequisites

Read access to the billing system tables is the hard dependency and the most common failure. system.billing.usage requires a metastore-admin grant chain — the skill detects a missing grant upfront and reports it, rather than failing mid-flow.

Databricks Premium or Enterprise workspace with Unity Catalog (system tables are UC-governed; not available on Standard).
Metastore-admin grant chain on the billing schema, granted by a metastore admin to the running principal:
- GRANT USE CATALOG ON CATALOG system TO <principal>
- GRANT USE SCHEMA ON SCHEMA system.billing TO <principal>
- GRANT SELECT ON TABLE system.billing.usage TO <principal>
- GRANT SELECT ON TABLE system.billing.list_prices TO <principal>
Compute system schema for config/utilization corroboration (same chain on system.compute): system.compute.clusters, system.compute.node_timeline.
Databricks CLI authenticated (databricks auth login, or DATABRICKS_HOST + DATABRICKS_TOKEN) and jq for parsing JSON tool output. The dollar queries run through the CLI Statement Execution API — UC enforces the grant chain above.
DATABRICKS_WAREHOUSE_ID env var set to a running SQL warehouse — every statement-execution call (Step 1's probe included) requires it.
databricks-workspace-mcp registered (its own PAT/U2M/M2M auth; PAT unsupported in Databricks-App deployment mode). It reads the live REST API, not system.*, so it needs no system-table grants. If it is absent the skill still produces dollar figures but cannot corroborate live config — it then accepts pasted config input.

Authentication. The CLI Statement Execution API uses DATABRICKS_HOST + DATABRICKS_TOKEN or databricks auth login; UC enforces the metastore grant chain on every system.* read. The custom databricks-workspace-mcp authenticates separately via its own PAT / U2M / M2M token. No secrets are hardcoded — all auth comes from the environment or the registered MCP server.

Run the upfront grant check before any analysis — see Step 1.

Instructions

The pipeline is detect → compute → rank → report. SQL detection runs through the CLI Statement Execution API; config corroboration runs on databricks-workspace-mcp; the dollar arithmetic runs in scripts/; deep knowledge loads from references/ on demand.

Step 1: Verify the Grant Chain (fail fast, not mid-flow)

Probe the billing tables before anything else. If the probe errors with a permission message, STOP and report the exact missing grant — do not continue into the leak scans. Requires DATABRICKS_WAREHOUSE_ID (a running SQL warehouse).

databricks api post /api/2.0/sql/statements --json '{
  "warehouse_id": "'"$DATABRICKS_WAREHOUSE_ID"'",
  "statement": "SELECT 1 FROM system.billing.usage LIMIT 1",
  "wait_timeout": "30s"
}' | jq -r '.status.state, .status.error.message // "ok"'

If state is not SUCCEEDED, load ${CLAUDE_SKILL_DIR}/references/system-tables-setup.md and report the missing grant chain to the user verbatim. Stop here.

Step 2: Pull the Spend Baseline

Establish the trailing-30-day total spend so every leak can be expressed as a share of a real number, and capture the window's MAX(usage_date) to stamp into the report. The price-window join (usage × list_prices.pricing.default, matched on sku_name AND usage_unit within the price-effective window, currency_code='USD') is the dollar primitive reused by every category query.

# The CLI does NOT expand ${VARS} inside a --json @file, so inject the warehouse
# id with jq at call time (the static template carries only wait_timeout + statement).
databricks api post /api/2.0/sql/statements --json "$(
  jq --arg wh "$DATABRICKS_WAREHOUSE_ID" '. + {warehouse_id: $wh}' \
    "${CLAUDE_SKILL_DIR}/scripts/sql/spend-baseline.sql.json"
)"

The canonical CTE and full per-category SQL live in ${CLAUDE_SKILL_DIR}/references/cost-leak-categories.md. Load it now — the four detection queries below all reference its priced CTE.

Step 3: Detect Leak 1 — Clusters That Never Auto-Terminate

Join priced All-Purpose usage to system.compute.clusters; flag clusters whose latest-change auto_termination_minutes = 0. Rank by 30-day idle spend. This is confirmed spend — money actually billed for idle compute.

SELECT p.usage_metadata.cluster_id AS cluster_id,
       COALESCE(c.cluster_name, 'unknown') AS cluster_name,
       c.auto_termination_minutes,
       ROUND(SUM(p.usd), 2) AS spend_30d_usd
FROM priced p
JOIN cluster_cfg c ON p.usage_metadata.cluster_id = c.cluster_id
WHERE p.billing_origin_product = 'ALL_PURPOSE'
  AND c.auto_termination_minutes = 0
GROUP BY p.usage_metadata.cluster_id, c.cluster_name, c.auto_termination_minutes
HAVING SUM(p.usd) > 0
ORDER BY spend_30d_usd DESC;

Corroborate each flagged cluster's live config with databricks-workspace-mcp clusters_get (confirm autotermination_minutes = 0 right now) and clusters_events (measure the idle gap between RUNNING and TERMINATING).

Step 4: Detect Leak 2 — Scheduled Jobs on All-Purpose Compute

The signature leak: a usage row with a job_id in usage_metadata AND billing_origin_product = 'ALL_PURPOSE' (~~$0.55/DBU) instead of JOBS_COMPUTE (~~$0.15/DBU). Re-price the same DBUs at the current Jobs rate to compute savings. This is confirmed savings — a deterministic re-pricing delta. The jobs_rate CTE is deduped to one USD rate per usage_unit so the join cannot fan out (see cost-leak-categories.md).

SELECT p.usage_metadata.job_id AS job_id,
       ROUND(SUM(p.usd), 2) AS spend_on_all_purpose_30d_usd,
       ROUND(SUM(p.usd) - SUM(p.usage_quantity * jr.jobs_unit_price), 2)
         AS potential_savings_30d_usd
FROM priced p
JOIN jobs_rate jr ON p.usage_unit = jr.usage_unit
WHERE p.billing_origin_product = 'ALL_PURPOSE'
  AND p.usage_metadata.job_id IS NOT NULL
GROUP BY p.usage_metadata.job_id
HAVING SUM(p.usd) - SUM(p.usage_quantity * jr.jobs_unit_price) > 0
ORDER BY potential_savings_30d_usd DESC;

Confirm the live compute type is All-Purpose (not Jobs) with clusters_list / clusters_get (cluster_source) before recommending the move. The job_id + ALL_PURPOSE billing signal is itself dollar-accurate; the REST check is belt-and-suspenders.

Step 5: Detect Leak 3 — Overprovisioned Clusters Idling Below Floor

Aggregate mean CPU from system.compute.node_timeline, join to 30-day spend, flag clusters burning real dollars at chronically low utilization (< 25%). This figure is an estimate (est_overprovision = spend × (1 − CPU%)), not billed waste — it is the one modeled number in the pipeline and is labeled est_* everywhere.

SELECT s.cluster_id,
       ROUND(u.avg_cpu_pct, 1) AS avg_cpu_pct,
       ROUND(s.spend_30d_usd, 2) AS spend_30d_usd,
       ROUND(s.spend_30d_usd * (1 - LEAST(u.avg_cpu_pct,100)/100.0), 2)
         AS est_overprovision_30d_usd
FROM spend s
JOIN util u ON s.cluster_id = u.cluster_id
WHERE u.avg_cpu_pct < 25 AND s.spend_30d_usd > 0
ORDER BY est_overprovision_30d_usd DESC;

Corroborate the configured floor with clusters_get (REST nested autoscale.min_workers / autoscale.max_workers); for the idle-pool variant use instance_pools_list (min_idle_instances + stats.idle_count) — pool waste is NOT a billing row.

Step 6: Detect Leak 4 — Photon Premium Without the Speedup

Photon is not a column on system.compute.clusters; it is billing-visible via the SKU. Isolate usage whose sku_name ILIKE '%PHOTON%' and surface the ~2× premium portion as the at-risk amount — money for review against actual runtime gain, not confirmed waste.

SELECT p.usage_metadata.cluster_id AS cluster_id,
       ROUND(SUM(p.usd), 2)        AS photon_spend_30d_usd,
       ROUND(SUM(p.usd) / 2.0, 2)  AS photon_premium_at_risk_30d_usd
FROM priced p
WHERE p.sku_name ILIKE '%PHOTON%'
  AND p.billing_origin_product IN ('ALL_PURPOSE','JOBS_COMPUTE')
  AND p.usage_metadata.cluster_id IS NOT NULL
GROUP BY p.usage_metadata.cluster_id
HAVING SUM(p.usd) > 0
ORDER BY photon_premium_at_risk_30d_usd DESC;

Confirm Photon is live and worth keeping with databricks-workspace-mcp clusters_get (REST runtime_engine — a config-plane field, not a system column); for DLT pipelines use pipelines_get (spec.photon / serverless / edition). See ${CLAUDE_SKILL_DIR}/references/dlt-tier-cost-tradeoffs.md when the leak touches DLT/serverless tiers.

Step 7: Compute, Rank, and Write the Report

Pass each category's query result to the deterministic ranker — the LLM does NOT do the arithmetic. Each leak object carries a kind field (confirmed / estimated / at-risk) so the renderer can split the headline into confirmed-recoverable vs estimated/at-risk-pending-review and stamp a Confidence column. The script sums per-category figures by kind, ranks descending by monthly dollar impact, annualizes the headline and #1 line, stamps the trailing-30-day window end date, and renders the CFO-grokkable report.

# Per-category results and the rendered report are RUNTIME outputs — they go to
# a working dir ($OUT), never the skill package. Steps 3–6 wrote leak-*.json here.
OUT="${OUT:-$(pwd)/cost-leak-out}" && mkdir -p "$OUT"
jq -s '.' "$OUT"/leak-*.json | \
  python3 "${CLAUDE_SKILL_DIR}/scripts/rank-and-report.py" \
    --monthly-spend 100000 \
    --window-end "$WINDOW_END_DATE" \
    --out "$OUT/cost-leak-report.md"

Use Glob to collect the per-category leak-*.json results, Write the rendered report, and Edit it if the user wants the headline spend rescaled. Render the output using the verbatim template in ${CLAUDE_SKILL_DIR}/references/cfo-output-format.md.

Output

A CFO-grokkable report file ($OUT/cost-leak-report.md in the working dir) leading with a split headline that never sums confirmed and unconfirmed dollars under one verb — ### A $<spend>/month workspace is burning **~$<confirmed>/month** (confirmed), plus up to **~$<at-risk>/month** pending review — each with its ~$<annualized>/year companion.
A trailing-30-day window stamp under the headline (Trailing 30 days ending <window-end>) so every figure has an explicit calendar window, not just a /month cadence label.
The ranked leak table with a Confidence column (# | Where it's leaking | $/month | Confidence | The fix), one row per category, ranked highest dollar impact first, $/month right-aligned, each fix a single config change. Root-cause cells use plain-business language — no raw DBU unit in the CFO-visible text (DBU detail stays in the per-leak detail artifacts).
The #1-line callout — the top leak annualized, named, with its confidence, and stated as fixed in one setting.
The assumed-vs-cited disclosure — only the workspace-spend input is assumed; on a live run confirmed figures are computed from system.billing.usage, while overprovision (estimated) and Photon premium (at-risk) are labeled as modeled.
Per-leak detail artifacts: idle-cluster list, all-purpose-job migration list with per-job savings, overprovisioned-cluster rightsizing list, Photon-premium at-risk list — each with the corroborating live config from databricks-workspace-mcp and the underlying $/DBU rates for engineers.

Error Handling

Error	Cause	Solution
`PERMISSION_DENIED` on `system.billing.usage`	Metastore-admin grant chain missing	Run Step 1; report the exact `GRANT USE CATALOG / USE SCHEMA / SELECT` chain from `system-tables-setup.md`. Stop, do not continue.
CLI not authenticated / token expired	No valid `DATABRICKS_HOST` + token	Re-run `databricks auth login`; verify with `databricks current-user me`.
Empty / unset `DATABRICKS_WAREHOUSE_ID`	Required warehouse for statement execution not set	Set `DATABRICKS_WAREHOUSE_ID` to a running SQL warehouse before Step 1.
`list_prices` join returns NULL `usd`	Custom/negotiated pricing not in `list_prices`, or `usage_unit` mismatch	Join on `sku_name` AND `usage_unit` within the price window with `currency_code='USD'`; if still NULL, use the customer's contracted rate card from `references/cost-leak-categories.md`.
Workspace MCP missing	Server not registered	Degrade gracefully: report it absent, run the dollar half, accept pasted config for corroboration. Never fail silently mid-flow.
`node_timeline` empty for a cluster	Serverless/short-lived compute, or monitoring lag	Skip Leak 3 for that cluster; note "utilization unavailable" rather than reporting $0 overprovision.
Untagged spend / no `cluster_name`	Clusters lack `CostCenter`/`Team` tags	Attribute by `cluster_id`; flag attribution as incomplete in the report footer.

Examples

Example 1: "Why is my Databricks bill high?"

Runs the full pipeline. The grant check passes, the four scans return rows, and the ranker emits the CFO report with a split, confidence-stamped headline:

### A $100K/month Databricks workspace is burning **~$19,000/month** (confirmed), plus up to **~$8,000/month** pending review

Trailing 30 days ending 2026-06-22. Confirmed ~$228K/year; up to ~$96K/year more pending review. Every line below is one config change.

| # | Where it's leaking | $/month | Confidence | The fix |
|---|---|--:|---|---|
| 1 | Clusters that never shut themselves off — paying around the clock for compute nobody is using | **$12,000** | Confirmed | Set auto-shutoff (e.g. 30 min) |
| 2 | Scheduled batch jobs running on the premium notebook tier — ~3.6× the batch rate for identical work | **$7,000** | Confirmed | Move job clusters to the batch tier |
| 3 | Clusters sized for peak, idling most of the time — typically 30–50% oversized | **$5,000** | Estimated | Turn on autoscaling, drop the floor |
| 4 | Paying a ~2× speed-engine premium on jobs that don't run faster | **$3,000** | At-risk | Turn off the speed engine where it adds no gain |

**The #1 line alone — idle clusters (confirmed) — is ~$144K/year, fixed in one setting.**

Example 2: Idle-Cluster Sweep

User asks "find idle clusters wasting money." The skill runs Step 3 only, joins the spend to clusters_get, and reports each auto_termination_minutes = 0 cluster with its 30-day idle spend and the live idle gap from clusters_events.

Example 3: All-Purpose-Job Rightsizing

User asks "are any jobs on the wrong compute?" Step 4 returns each job_id running on All-Purpose with potential_savings_30d_usd, corroborated by clusters_get confirming cluster_source is not JOB — the single fix is "move to Jobs Compute."

Resources

${CLAUDE_SKILL_DIR}/references/cost-leak-categories.md — the four leak categories: definition, real detection SQL, FinOps root cause, remediation.
${CLAUDE_SKILL_DIR}/references/cfo-output-format.md — verbatim CFO report template + the 90-second-skim rules.
${CLAUDE_SKILL_DIR}/references/system-tables-setup.md — metastore-admin grant chain + access verification.
${CLAUDE_SKILL_DIR}/references/dlt-tier-cost-tradeoffs.md — DLT / serverless / Photon cost-tier encyclopedia, loaded on demand.
Databricks system tables (billing)
Databricks list_prices reference

When it activates

Automatically, when a task matches. Use when a user asks why their Databricks bill is high, wants to find cost leaks / wasted DBUs / idle clusters, or needs a FinOps cost report. Trigger with "databricks cost", "why is my databricks bill", "find wasted spend", "cost leak".

About this skill

Automatic

Data Engineer pack

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

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

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

jeremylongshore / claude-code-plugins-plus-skills

skills/.curated/databricks-cost-leak-hunter/SKILL.md

View on GitHub ↗

License MITImported 10/23/2026Version v1Bundled 5 files · scripts

Illustrative Want to see what jeremylongshore/databricks-cost-leak-hunter 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

Databricks Cost Leak Hunter

Overview

Prerequisites

Instructions

Step 1: Verify the Grant Chain (fail fast, not mid-flow)

Step 2: Pull the Spend Baseline

Step 3: Detect Leak 1 — Clusters That Never Auto-Terminate

Step 4: Detect Leak 2 — Scheduled Jobs on All-Purpose Compute

Step 5: Detect Leak 3 — Overprovisioned Clusters Idling Below Floor

Step 6: Detect Leak 4 — Photon Premium Without the Speedup

Step 7: Compute, Rank, and Write the Report

Output

Error Handling

Examples

Example 1: "Why is my Databricks bill high?"

Example 2: Idle-Cluster Sweep

Example 3: All-Purpose-Job Rightsizing

Resources

When it activates

SkillSafety

Source

Related skills

More in the registry