Cloud Spend Advisor skill for Claude and Codex
A prioritised cloud cost-cutting plan for AWS, Azure or Google Cloud that holds performance, reliability and security steady, with savings estimates.
Install it once and Claude or Codex runs it for you. When your request matches, your AI loads the skill, asks for anything it is missing, and works through the template itself. Prefer to fill it in by hand? See the full prompt template.
Download the skill (.zip) · View SKILL.md
When it is used
Use when an engineering or finance lead wants a ranked list of ways to cut cloud spend while keeping performance and reliability intact. Do not use for a cloud migration plan or a security audit of the infrastructure. Related topics: cloud, devops, finance.
What it will ask you for
- Providers (choose any: AWS, Azure, Google Cloud, Other)
- If other, which (free text, e.g. "Oracle Cloud or Alibaba Cloud")
- Monthly spend (free text, e.g. "For example 15000 USD")
- Workloads (choose any: Web applications, APIs, Data processing, Machine learning, Databases, Batch jobs, Internal tools, IoT, Other)
- Services in use (longer free text)
- Performance needs (longer free text)
- Reliability targets (longer free text)
- Security and compliance (longer free text)
- Main cost concerns (choose any: Compute, Storage, Networking, Databases, Data transfer, Licensing, Managed services, Idle resources, Unused resources)
- Known pain points (longer free text)
- Priority areas (longer free text)
- Time horizon (choose one: Quick wins within a month, Within three months, Within a year, Beyond a year, All horizons)
- Detail level (choose one: High-level overview, Actionable steps with impact, Detailed technical guidance)
- Layout (choose one: Bulleted list, Numbered steps, Table, Executive summary with breakdown)
- Estimate savings (choose one: Yes, No)
- Estimate performance impact (choose one: Yes, No)
- Focus on (choose any: Rightsizing, Commitment discounts, Spot instances, Serverless, Storage tiering, Network optimisation, Database optimisation, Automation, Cost governance)
How it works
- Map the spend, workloads and services to the likeliest sources of waste for the stated concerns.
- Screen out any change that would breach the performance, reliability or compliance needs.
- Rank the remaining recommendations by saving against effort and risk, within the chosen time horizon.
- For each, state the change, the expected saving if requested, the performance risk if requested and a way to verify it.
- Present the list in the chosen layout and detail level, labelling estimates as estimates.
What you get
A ranked list of recommendations in the chosen layout, each with change, estimated saving, risk and verification, plus a note on assumptions.
What it watches out for
- Recommending reserved capacity for a workload that may shrink, which locks in cost.
- Rightsizing on average usage while ignoring peaks, which causes slowdowns.
- Quoting precise savings from a single monthly spend figure with no usage data.
How to install
Claude.ai and the Claude desktop app
- Download the skill .zip.
- Open Settings, then Capabilities, and find Skills.
- Choose Upload skill and select the .zip.
- Ask Claude for help with the task. It loads the skill when your request matches.
Claude Code
- Download the skill .zip.
- Unzip it into ~/.claude/skills/ (or .claude/skills/ inside a project).
- Start a new session. Claude Code loads the skill when your request matches.
Codex
- Download the skill .zip.
- Unzip it into ~/.agents/skills/ (or .agents/skills/ inside a project).
- Codex picks it up on its own. Type $ and the skill name to use it, or just ask for the task.
Cursor
- Download the Cursor rule (.mdc).
- Save it in .cursor/rules/ inside your project.
- Cursor's agent applies the rule when your request matches its description.
AGENTS.md (Codex, Cursor, and other coding agents)
- Copy the AGENTS.md section.
- Paste it into AGENTS.md at the root of your repository, creating the file if needed.
- Coding agents that read AGENTS.md follow it when your request fits.