\n \n

Cloud FinOps skill - Watch Video

\n \n \n $\"Demo$ \n \n

\n\n- \"We're spending $40K/month on AWS Bedrock and have no idea which features are driving it. Where do we start?\"\n- \"How do I calculate the break-even utilisation rate for provisioned throughput - and should we choose Azure OpenAI PTUs or Bedrock provisioned capacity for 500K requests/day?\"\n- \"Our monthly bill jumped from $12K to $38K after a developer enabled Fast mode in Claude Code. How do I get this under control?\"\n- \"Should we self-host Llama 4 on rented H100s instead of paying per token - and what hidden costs do TCO calculators miss?\"\n- \"We have $80K/month in EC2. Reserved Instances or Savings Plans - and what quick wins come first?\"\n- \"Our client wants separate AWS invoices per business unit. Their AWS contact suggested Cost Categories - is that right?\"\n- \"We're migrating from EA to MCA - what FinOps work do we need to do before the switch?\"\n- \"Which VMs run for nothing, and which runbook finds them?\"\n- \"We need to start reporting our cloud carbon emissions - where do we begin?\"\n\n---\n\n## Directory structure\n\n```\ncloud-finops-skills/\n├── README.md <- This file\n├── INSTALLATION.md <- Per-tool setup, troubleshooting, API loader\n├── CLAUDE.md / AGENTS.md <- Project context for AI assistants and contributors\n├── llms.txt <- LLM discovery index (cross-agent)\n├── install.sh <- Cross-tool installer (12 targets)\n├── mcp_server/ <- cloud-finops-mcp PyPI package\n└── skills/cloud-finops/ <- The skill - install this folder\n ├── SKILL.md <- Entry point + per-file routing catalogue\n ├── POWER.md <- Kiro IDE entry point (same references)\n ├── references/ <- The reference library, one file per domain\n └── playbooks/ <- Named-pattern runbooks (~3-8 KB each) + catalogue\n```\n\n---\n\n## MCP server (cross-tool, search-style retrieval)\n\nHosted (nothing to install) or from PyPI - both paths are in the\n[install table](#install-in-5-seconds) above. Also listed on the\n[MCP Registry](https://registry.modelcontextprotocol.io/v0/servers?search=finops) as\n`io.github.OptimNow/cloud-finops`, and on\n[PyPI](https://pypi.org/project/cloud-finops-mcp/).\n\nSix read-only tools across two surfaces. The split is deliberate: the two content\ntypes have different shapes, and different questions attached to them.\n\n**References** - the long-form provider and discipline files. Reach for these for\nbilling mechanics, commitment strategy, allocation methodology, or any reasoning that\nspans patterns.\n\n| Tool | What it answers |\n|---|---|\n| `list_references()` | What guidance exists? The catalogue with its FinOps Framework facets and an `approx_tokens` size hint per file |\n| `get_reference(name, section?)` | One guide - mechanics, decision rules, worked examples. Whole, or a single H2/H3 section when the question is narrower than the file |\n| `find_references(domain?, capability?, phase?, persona?, maturity?, persona_primary_only?)` | \"How should we size Savings Plans?\" \"What must be true before chargeback?\" - routes a FinOps question to the guides that serve it (`persona_primary_only` cuts to the primary audience) |\n\n**Playbooks** - small named-pattern runbooks, one waste pattern each. Reach for these\nfor \"how do I detect and fix this specific thing\".\n\n| Tool | What it answers |\n|---|---|\n| `list_playbooks()` | What cloud waste can we hunt with a ready-made runbook? |\n| `get_playbook(name)` | The step-by-step runbook: symptoms, detection queries, fix, anti-pattern |\n| `find_playbooks(scope?, service?, waste_category?, confidence?)` | \"Which VMs run for nothing?\" \"Why is the NAT bill so high?\" \"Which of my RIs are about to expire?\" - finds the runbook for a specific waste suspicion. The server cannot see your account; the runbook's detection query is the answer it hands over |\n\nBoth listings carry `approx_tokens` per entry, because the references vary by more\nthan tenfold - roughly 2K tokens for the smallest, over 25K for the provider pattern\ncatalogues. The catalogues are enumerated lists, so an agent that wants one pattern\nfamily passes `section` (`get_reference(\"finops-aws-patterns\", section=\"storage\")`)\nand pays for that section instead of the whole file. Matching is case-insensitive and\npartial; a phrase that matches no heading returns the file's available headings rather\nthan silently falling back to the full body.\n\nThe faceted queries are the reason this is a server and not just a folder of markdown:\nevery file carries YAML frontmatter mapping it to a FinOps Framework capability, phase,\npersona and maturity gate, and a client that only fetches files cannot filter on any\nof it.\n\nOn hosts that support MCP Apps (SEP-1865), the tool results render as interactive\nwidgets - a playbook explorer with facet filters and a coverage matrix, a playbook\nviewer with copyable detection queries and a checkable fix list, and a reference\nbrowser with a reading panel. Hosts without MCP Apps support get the plain results;\nnothing about the tools changes. Details in\n[mcp_server/README.md](https://github.com/OptimNow/cloud-finops-skills/blob/main/mcp_server/README.md).\n\n---\n\n## This skill is actively maintained\n\nThis is a living repository. Reference files are refreshed twice a month (around the\n1st and the 15th), driven by an automated scan of around 30 data sources - cloud\nprovider pricing pages, release notes, billing changelogs, and FinOps community\npublications. Changes are reviewed before being applied, so the content reflects\nverified updates rather than raw feed output.\n\nAI cost management is moving particularly fast - new model releases, capacity options,\nand billing mechanics appear every few weeks. Watch or star this repo to be notified\nwhen updates are published.\n\n---\n\n## Contributing\n\nPractitioner experience is the highest-value contribution. Frameworks and vendor docs\nare already public; what is rare is \"we tried X in production, this is what actually\nbilled\". Corrections to billing mechanics, new or improved playbooks, real-world\ncounter-examples, and adversarial review of the recommendations are all welcome - the\nrepo is opinionated, and it should also be falsifiable.\n\nThe full guide - contribution types, process, conventions, and what we push back on -\nis in [CONTRIBUTING.md](https://github.com/OptimNow/cloud-finops-skills/blob/main/CONTRIBUTING.md). One check before anything else: if your\nchange names another OptimNow tool (an MCP tool name, an endpoint URL, a provenance\nfield), read [DEPENDENCIES.md](https://github.com/OptimNow/cloud-finops-skills/blob/main/DEPENDENCIES.md) first - most cross-repo breakage\nhere is documentation drift that no CI check catches.\n\n---\n\n## Adapting this skill for your organisation\n\nFork this repository and customise the reference files for your organisation's context:\nyour cloud stack, your internal policies, your tag taxonomy, your preferred methodology.\n\nA fork gives you a stable base that you can pull upstream updates into at your own pace,\nwithout overwriting your customisations. Typical customisations include:\n\n- Adding organisation-specific tag requirements to `finops-tagging.md`\n- Replacing generic pricing examples with your negotiated rates\n- Adding reference files for internal tools or platforms not covered here\n- Adjusting the methodology file to reflect your team's own approach\n\n---\n\n## About OptimNow\n\nOptimNow is a boutique FinOps consultancy helping organisations connect cloud and AI\nspend to measurable business value. Based in France with European reach.\n\n- Website: [optimnow.io](https://optimnow.io)\n- LinkedIn: [OptimNow](https://linkedin.com/company/optimnow)\n- GitHub: [github.com/OptimNow](https://github.com/OptimNow)\n\n**Open-source tools built by OptimNow:**\n\n| Tool | What it does |\n|---|---|\n| [OptimToken](https://optimtoken.optimnow.io) | Compare what 250+ models cost per request, with caching and batch factored in, plus compute instance rates across seven clouds. Also available as an MCP connector - this skill routes price questions here |\n| [AI ROI Calculator](https://airoicalculator.optimnow.io) | Whether an AI project pays for itself: three-layer cost model, payback, break-even, sensitivity. Also an [MCP server](https://github.com/OptimNow/ai-roi-calculator-mcp) |\n| [AI Cost Readiness Assessment](https://aicostsfinops.optimnow.io) | Where your organisation stands on AI cost management |\n| [MCP for Tagging](https://github.com/OptimNow/finops-mcp) | Tag governance automation |\n| [FinOps Maturity Assessment](https://optimnow.io) | Crawl / Walk / Run positioning |\n\n---\n\n## Acknowledgements\n\nThis skill incorporates content derived from the following sources:\n\n- **[FinOps Foundation](https://www.finops.org/)** - framework definitions, capability\n descriptions, and maturity model structure are based on the FinOps Framework.\n- **[Point Five](https://www.pointfive.co)** - cloud optimisation recommendations\n informed several provider-specific best practices and quick-win patterns.\n- **[Tokenomics Foundation](https://www.tokeneconomics.com/)** - the token complexity\n classes in the agentic FinOps reference are adapted from *Big-T Notation* by\n Dan Neff (Adobe), published by the Tokenomics Foundation under\n [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/).\n\nAll referenced content has been adapted with additional context from OptimNow's\nconsulting delivery experience. Any errors or opinionated interpretations are our own.\n\nThis skill is independently maintained and is not affiliated with or endorsed by the\nFinOps Foundation.\n\n---\n\n## License\n\nLicensed under [CC BY-SA 4.0](https://creativecommons.org/licenses/by-sa/4.0/).\nSee [LICENSE.md](https://github.com/OptimNow/cloud-finops-skills/blob/main/LICENSE.md).\n\nYou are free to use, adapt, and redistribute this skill - including for commercial\npurposes - as long as you credit OptimNow and share any derivatives under the same license.\n","changelogMarkdown","videoKey","requiredSecretsJson","publishedAt","priceCents","trialDays",0,"upstream",{"_110":111,"_112":113,"_114":115,"_116":117,"_118":119,"_120":121},"requiredSecrets",[],"tags",[95,96,97,98,99,100,101,102,103,104,105,106,107,108,109],"assets",[],"stats",{"_93":82,"_94":82},"ratingAvg","reviewCount","agent-skills","ai-cost-management","aws","azure","chargeback","claude-code","cloud-cost-optimization","cloud-financial-management","cloud-governance","databricks","finops","finops-foundation","gcp","llm-cost","skill-md","topics",[95,96,97,98,99,100,101,102,103,104,105,106,107,108,127,128,129,109,130,131],"license","NOASSERTION","language","Python","registryName","io.github.OptimNow/cloud-finops","homepage","https://www.optimnow.io","packages",[122],{"_123":124,"_125":126,"_72":73},"registryType","pypi","identifier","cloud-finops-mcp","mcp","mcp-server","model-context-protocol","snowflake","tagging","ENV",{"_134":135,"_136":137,"_138":139},"ENVIRONMENT","production","TURNSTILE_SITE_KEY","0x4AAAAAAB8-rEI2gNDf_QhU","SENTRY_DSN","https://45fd1229f78089d5c3ee78be128030e1@o4510583261757440.ingest.de.sentry.io/4510584253448272"]

Cloud FinOps Skill & MCP