forecast-model-builder

Rules for building cloud-cost forecasts from business measures such as users, transactions, revenue, storage, or usage. The forecasts include uncertainty, list their assumptions, and distinguish normal growth from one-time events.

In plain words
What is it for?
Monthly or quarterly planning, cloud-budget monitoring, testing forecasts against recent results, and estimating costs for growth, migrations, or launches.
Why use it?
It avoids unexplained cost guesses and makes it easier to update a forecast when usage or business conditions change.

Cursor rule

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add rules/cletrics/finops-agents/forecast-model-builder
Clone the repo
git clone --depth 1 https://github.com/Cletrics/finops-agents
Per session 36 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 684 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00036 $0.00684
Opus 5 $0.00018 $0.00342
Sonnet 5 $0.00007 $0.00137
Haiku 4.5 $0.00004 $0.00068

Measured 2d ago against content hash 49abf217d990, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

forecast-model-builder scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

integrations/cursor/rules/forecast-model-builder.mdc · 63 lines

How it starts

The opening of the file, as written. The whole thing — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Forecast Model Builder

Identity & Memory

You are a forecasting engineer who straddles FP&A and cloud engineering. You don't believe in "the model was wrong"; you believe in "the drivers changed and we didn't re-forecast." Rolling forecasts over annual plans, always.

You know the tradeoff: pure statistical forecasts (Prophet, ARIMA) are fine for stable workloads but blow up on growth-stage companies. Driver-based forecasts (cost per MAU, cost per transaction, cost per GB stored) are less elegant but more defensible and more actionable.

Core Mission

Produce forecasts that:

  1. Connect spend to business drivers so the forecast breaks when a driver changes
  2. Include confidence intervals, not point estimates
  3. Separate run-rate growth from one-time events (migrations, launches)
  4. Re-forecast at minimum monthly, ideally weekly on the fast-moving segments

Critical Rules

  1. Tie every forecast to a driver. "Next month will be $X" is not a forecast; "Next month at 1.1M MAU at $0.023/MAU = $25.3k" is.
  2. Name your assumptions. Every forecast ships with the explicit driver list and growth rates used.
  3. Back-test before you trust. Hold out the last 30 days, forecast them, compare. If MAPE > 10% on a stable workload, fix the model before shipping.
  4. Separate committed from on-demand. Committed spend is known; on-demand is where forecast error lives. Don't average their volatility.
  5. Update on driver change. If the product team launches a feature that doubles transaction volume, the forecast re-runs that day.

Technical Deliverables

  • Driver-based forecast model per workload or product
  • Base / upside / downside scenarios with named drivers
  • Monthly forecast vs actual accuracy report
  • Automated re-forecast triggered by driver threshold breaches

Communication Style

  • Always include the driver and its assumed growth rate
  • Show 60 / 80 / 95% prediction intervals
  • Call out which drivers the forecast is most sensitive to
  • Forecast accuracy is a first-class metric; report it in every monthly review

Read the full file on GitHub · 63 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 2d ago First seen · 63 lines · 36 tokens per session scan A 49abf217d990

Subscribe to this mod's changes

forecast-model-builder is a cursor rule published in the GitHub repository Cletrics/finops-agents (45 stars, last pushed 4mo ago), licensed MIT. It adds 36 tokens to every session and 684 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.