AI Prompt Library
Put AI to work using prompts that fit seamlessly into your workflows, whether in-app with our Lux agent, or through your preferred AI platform. Bulk create and manage items, quickly report and analyze data, and handle everything from budget management to board-ready cost summaries with ease. Get started with our MCP Server.
Some prompts require additional MCP connections (Pendo, GitHub, GitLab, AWS, etc).
FinOps Crawl
What are our top cost-driving services right now?
What are the top 5 cloud services driving spend across all providers today? Break it down by provider and show me the month-over-month trend for each.
FinOps Crawl
Are we over or under budget this month?
Show me budget vs. actual spend for this month across all accounts. Which OUs or teams are driving the biggest variances, and are any budgets at risk of breaching before month end?
FinOps Crawl
Which cloud accounts are costing the most?
Which accounts across my cloud providers are spending the most right now, and what are they used for? Group by OU and show me the top 10.
FinOps Crawl
Multi-cloud spend summary for last month
Give me total cloud spend for the most recently completed month, broken down by provider and top 5 services per provider. Flag anything that moved more than 15% month-over-month.
FinOps Walk
Detect and explain unusual spend spikes from the past 7 days
Are there any accounts, OUs, or services where spend exceeded expected thresholds in the past 7 days? For each anomaly, tell me when it started, how much above baseline it is, and whether Kion has any associated budget alerts or compliance findings that overlap with the affected resources.
FinOps Walk
Where are we wasting money on idle or orphaned resources?
Identify resources that look idle or orphaned across all cloud accounts — unattached storage volumes, stopped instances still incurring charges, unused load balancers, and non-production workloads running around the clock. Estimate the monthly waste for each category.
FinOps Walk
Show me underutilized compute that could be rightsized
Show me compute resources across AWS and Azure that are consistently underutilized. Recommend rightsizing actions and estimate the monthly savings for each.
FinOps Walk
How much are we spending on non-prod infrastructure after hours?
Identify non-production workloads that are running outside of business hours — Monday through Friday, 8am to 8pm local time. Show the services and accounts involved, calculate how much that off-hours spend adds up to monthly, and model the savings if those workloads ran on a schedule instead. Flag any resources that have been exempt from tagging or scheduling policies.
FinOps Walk
What compliance policies are generating the most violations?
Which Cloud Custodian policies in Kion are generating the most active violations right now? Rank them by violation count, show me which accounts and resource types are most affected, and tell me whether any of these violations overlap with accounts that are also over budget or showing spend anomalies.
FinOps Run
Build a weekly cost and governance digest for the team
Build a weekly digest covering: top 3 spend changes vs. last week, any anomalies detected, current commitment utilization, active budget alerts, and open compliance findings with policy violations. Format it as a concise summary ready to post in a team Slack channel. This is designed to run on a recurring schedule.
FinOps Run
Triage cost anomalies using Slack conversations
Pull a Kion spend report for all accounts and cloud services over the past 28 days. For each account/service: Baseline: days 8–28 (3-week daily average) Flag if: past 7 days exceeded baseline by >20% on ≥2 consecutive days Ignore: accounts spending <$50/day For each anomaly, capture: Account name + ID, cloud service/resource type Anomaly window (start/end), peak overage %, estimated excess cost ($) Then search Slack (public + private) using the account name, service name, and relevant team/cost center. Look for messages within ±48 hours of the anomaly start — prioritize threads mentioning deployments, incidents, autoscaling, load tests, migrations, or budget approvals. Summarize each finding with: Event: what spiked, when, by how much Root cause: Known / Likely known / Unexplained Evidence: relevant Slack quote or link, or "No Slack activity found" Awareness: who knew at the time Disposition: Close / Monitor / Escalate Sort all anomalies by excess cost (descending). Open with a one-line executive summary: total anomalies, total excess spend, and escalation count.
FinOps Run
Triage cost anomalies using CloudWatch
Using AWS CloudWatch, pull metric data across all monitored resources and namespaces for the past 28 days. For common metrics (CPUUtilization, NetworkIn/Out, RequestCount, Invocations, ReadIOPS, WriteIOPS): Baseline: days 8–28 (3-week average) Flag if: the past 7 days exceeded baseline by >20% on ≥2 consecutive data points Ignore: low-signal resources (CPU <5%, requests <100/day) For each anomaly, capture: Account ID + alias, resource ARN/name, metric + namespace Anomaly window (start/end), peak deviation %, estimated cost impact Summarize each finding with: Event: what spiked, where, when, by how much Root cause: Known / Likely known / Unexplained Awareness: who knew at the time Disposition: Close / Monitor / Escalate
FinOps Run
What is our cost per customer, and how is it trending?
Calculate our average infrastructure cost per active customer over the past 6 months. Break it down by cloud provider and major service category. Flag any months where cost grew faster than customer count and identify which services drove the divergence.
FinOps Run
Surface product usage patterns that explain cost growth
Using Pendo, pull active account counts, feature adoption rates, and session volume trends for the past 90 days. Then pull the corresponding cloud spend trend from Kion for the same period, broken down by service. Correlate the two — where is cost growing in proportion to usage (expected), and where is cost growing faster than usage (worth investigating)? Identify the services or accounts where spend efficiency is degrading and suggest whether the cause is infrastructure behavior, feature usage patterns, or account tier mix.
FinOps Run
Board-ready cloud governance and cost summary
Produce an executive summary suitable for a board or QBR: total cloud spend trend across the last 3 quarters, cost efficiency vs. revenue growth, top 3 savings opportunities with estimated impact, commitment coverage status, compliance posture summary, and a 2-quarter forecast with confidence ranges.
FinOps Run
Flag compliance violations with no open [GitLab or GitHub] issue
Pull the current list of open compliance findings from Kion — policy violations, non-compliant resources, and failed checks. Then search GitLab for open issues or merge requests that reference the same resource names, account IDs, or policy names. For each compliance finding that has no corresponding GitLab or GitHub issue, draft a summary of the violation with enough context for an engineering team to triage it. Flag any findings that have been open for more than 30 days with no associated GitLab activity.