메뉴
Produce a one-paragraph executive summary on a FinOps topic, in house style. Triggers on phrases like "exec digest", "CFO summary", "one-paragraph summary for leadership", "write this up for the exec readout". Lead with dollars, no jargon, action-oriented close.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | exec-digest |
| description | Produce a one-paragraph executive summary on a FinOps topic, in house style. Triggers on phrases like "exec digest", "CFO summary", "one-paragraph summary for leadership", "write this up for the exec readout". Lead with dollars, no jargon, action-oriented close. |
When invoked, produce a single tight paragraph suitable for an executive readout, weekly digest, or the top of a longer doc that a busy executive will only skim. Always one paragraph. Always under 100 words. Always in house style.
These are what makes a paragraph feel like ours and not a generic AI blob. Apply all of them, every time.
[paragraph]
---
Word count: N
Audience tested for: [CFO / CTO / general exec]
Action implied: [one line — what the reader is being asked to do or notice]
The metadata footer is for the practitioner's eyes, not the digest. They'll strip it before sending.
Good action-oriented closes:
Bad closes:
Before delivering, run yourself through:
If any answer is no, rewrite.
Investigate a cost anomaly end-to-end. Triggers on phrases like "cost spike", "unexpected spend", "anomaly", "what's driving this cost". Pulls billing data, identifies the top contributing SKUs/projects/labels, drafts a hypothesis about cause, and produces both the engineering Slack message and the CFO digest paragraph in house style.
Report on current utilization across all active commitments (CUDs, RIs, Savings Plans). Triggers on phrases like "commitment utilization", "are we using our CUDs", "RI report", "savings plan check", "commitment portfolio". Computes current utilization, days remaining, projected over- or under-use, and the dollar impact of any gap.
Generate a standardized writeup after a cost anomaly has been resolved. Triggers on phrases like "postmortem for the X spike", "writeup for the anomaly", "document what happened with X". Produces a structured document covering timeline, root cause, dollar impact, what was changed, and prevention — so postmortems are comparable across the team over time.
Prepare the FinOps section of the monthly business review. Triggers on phrases like "MBR prep", "monthly business review", "monthly cost review", "I need to put together my MBR". Pulls last month's spend by service and team, computes trends vs the prior 3 months, drafts the narrative that goes in the deck, and surfaces the anomalies leadership will ask about.
Save and manage prompts that worked well. Triggers on phrases like "save this prompt", "add to prompt library", "this is a keeper", "remember this prompt". Captures the prompt with metadata (use case, model version, what it's good at, what to watch for) so it can be retrieved and reused later.
Generate ticket-ready right-sizing actions for a service, workload, or team. Triggers on phrases like "right-size", "rightsizing", "are we over-provisioned", "what can we shrink", "utilization opportunities". Pulls utilization data, generates concrete recommendations with dollar impact estimates, formatted so the practitioner can paste each one into a ticket and assign it.