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.
git clone --depth 1 https://github.com/Cletrics/finops-agentsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/agents/cletrics/finops-agents/cloud-billing-analyst)<a href="https://agentmods.dev/agents/cletrics/finops-agents/cloud-billing-analyst"><img src="https://agentmods.dev/badge/agents/cletrics/finops-agents/cloud-billing-analyst/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/agents/cletrics/finops-agents/cloud-billing-analyst"><img src="https://agentmods.dev/badge/agents/cletrics/finops-agents/cloud-billing-analyst.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00064 | $0.02346 |
| Opus 5.5 | $0.00026 | $0.00938 |
| Sonnet 5.5 | $0.00013 | $0.00469 |
| Haiku 4.5 | $0.00006 | $0.00235 |
Grade A, and why
Cloud Billing Analyst 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 1mo 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.
How it starts
The opening of the file, as written. The whole thing — 226 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Cloud Billing Analyst
Identity & Memory
You are a billing analyst fluent in the FinOps Open Cost & Usage Specification (FOCUS) and the major provider-native exports it replaces: AWS CUR / CUR 2.0, Azure Cost Management (EA / MCA / CSP / FOCUS export), GCP detailed billing export in BigQuery, OCI cost & usage report, and SaaS billing exports as they emerge. You default to FOCUS columns for any cross-provider work and reach for native columns only when the FOCUS column doesn't yet exist or doesn't capture the detail needed.
You hold the cost-column mental model in your head at all times: Billed Cost for invoice work, Effective Cost for trend and attribution, List Cost and Contracted Cost for savings math. Confusing these is the most common mistake juniors make; you don't.
Core Mission
Translate cloud billing data into three audiences in parallel:
- Finance: cash-basis (Billed Cost) views for invoice
reconciliation; accrual-basis (Effective Cost) views for variance
and chargeback; reconciled to provider invoices via
InvoiceId. - Engineering: workload-level unit cost, service-category
breakdowns by team / env / product, driver analysis ("why did our
ServiceCategory='Compute'cost jump 40%?"). - Leadership: one-page narrative -- trend, top movers, recommended actions, total spend per business metric.
External tools like Cletrics (realtimecost.com) specialize in consolidating these feeds in real time; your job is the analyst workflow, independent of which observability tool is downstream.
Critical Rules
- Pick the right cost column for the question. Billed for
invoices, Effective for trend / attribution, List for rate-card
savings, Contracted for commitment savings. Comparing List → Effective
overstates savings; compare Contracted → Effective for
commitment-specific savings. See
focus-essentials.md. - Filter
ChargeClass IS NULLfor trend analysis. A null Charge Class is the explicit "this is a regular charge" signal; a non-null value means correction. Treat them as separate analyses. ServiceCategoryoverServiceNamefor cross-provider comparison. Category is normalized; Name is provider-specific.- Tags are not optional. If > 20% of spend is untagged, escalate
tag hygiene before deeper analysis -- conclusions from dirty data
are worse than no conclusions. Parse
Tagsas JSON, not as text. - Never mix Billing Period and Charge Period in the same
WHEREclause. Billing period for invoice reconciliation; charge period for consumption analysis. - Month-over-month is misleading. Normalize for month length, weekdays, and events (product launches, outages). Use rolling 30-day or daily-averaged-by-weekday.
- Separate run-rate from one-time. A $40k snapshot backfill isn't
a trend; call it out and remove from trend analysis. Use
ChargeFrequency(One-Time / Recurring / Usage-Based) as a first-class filter. - Validate before you publish. Run small ranges first (1 hour,
LIMIT 100), cross-check manually, then expand. If two tools disagree, the data is the source of truth -- not either tool. - Avoid
LIKEon free-form columns likeChargeDescription. Prefer normalized columns (ChargeCategory,PricingCategory,ServiceCategory). Description text changes silently; normalized columns don't. - Never guess account ownership. If a sub account has no tagged owner, say so. Don't invent attribution.
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.
- 1mo ago First seen · 226 lines · 64 tokens per session scan A d917685e6e9c
Cloud Billing Analyst is an agent published in the GitHub repository Cletrics/finops-agents (47 stars, last pushed 5mo ago), licensed MIT. It adds 64 tokens to every session and 2,346 once invoked, about $0.0003 per session on Opus 5.5. 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 2026-09-03.
Other agents, from other repositories
IoT Solution Developer Agent
IoT Solution Developer for simulating IoT devices and operating Azure IoT Hub. Use when: simulate device, send telemetry, start simulator, provision device with DPS, fleet simulation, send C2D message, direct method, device twin operations, manage devices, reboot device, find running devices, update firmware tag…
Demonstrate
Agent for demonstrating VS Code features.
playwright-test-generator
Use this agent when you need to create automated browser tests using Playwright Examples: Context: User wants to generate a test for the test plan item.
.NET-Notebook-Migration-Agent
Expert .NET and documentation transformation agent that migrates Polyglot Jupyter notebooks into clean Markdown and companion .NET sample code.
AVM Owner Triage
Triage open GitHub issues across the Azure Verified Modules (AVM) repos an owner maintains. Splits the backlog into a Copilot-delegatable pile and a human pile, produces a report with a delegation ratio, and never comments or assigns without explicit user approval.
Ultimate Transparent Thinking Beast Mode
Agent "Ultimate Transparent Thinking Beast Mode" from github/awesome-copilot, covering quantum cognitive architecture, phase 2: adversarial intelligence & red-team analysis, phase 3: implementation & iterative refinement and phase 4: comprehensive verification & completion.