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.
npx skills add getnable/finopsmcp --skill check-ai-budgetgit clone --depth 1 https://github.com/getnable/finopsmcpWrote 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/skills/getnable/finopsmcp/check-ai-budget)<a href="https://agentmods.dev/skills/getnable/finopsmcp/check-ai-budget"><img src="https://agentmods.dev/badge/skills/getnable/finopsmcp/check-ai-budget/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/skills/getnable/finopsmcp/check-ai-budget"><img src="https://agentmods.dev/badge/skills/getnable/finopsmcp/check-ai-budget.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 | $0.00053 | $0.00441 |
| Opus 5.5 | $0.00021 | $0.00176 |
| Sonnet 5.5 | $0.00011 | $0.00088 |
| Haiku 4.5 | $0.00005 | $0.00044 |
Grade A, and why
check-ai-budget 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 today.
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.
What it actually says
Check the AI budget before a long task
Your own tokens cost money, and a long task can spend in an hour what a normal day spends. nable tracks this agent's spend against the budget the person set, on this machine.
Before a long task:
- Estimate the tokens the task will take. A rough guide: reading a large file is 10 to 50 thousand tokens, a change across 20 files with tests is 500 thousand to 2 million, a long autonomous loop can pass 5 million.
- Call the
check_ai_budgettool withestimated_next_tokensset to that estimate (or runnable ai-budget --jsonwhere there is no MCP server). - Read the verdict and relay it in one sentence, with the headroom it
reports:
ok: go ahead.warn: tell the person what is left and what the task will likely use, and offer a smaller first step. Continue only if they say so.over: stop before starting and ask. Do not start a reduced version on your own.
- Midway through a task that is running long, check again.
Rules:
- The verdict is advice and the person decides. Never change the budget
yourself (
set_ai_budgetis theirs to call), and never split a task into smaller sessions to stay under a cap. - Dollar figures are list-price estimates from token counts, not the bill. Say so if you quote one.
- When you write code that calls an LLM API, tag each call with
feature:<name>and, where there is one,customer:<id>(a LiteLLM request tag or a Langfuse trace tag), so the spend can be attributed.
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.
- today First seen · 38 lines · 53 tokens per session scan A 1ab1afd8fbb0
check-ai-budget is a skill published in the GitHub repository getnable/finopsmcp (19 stars, last pushed today), licensed Apache-2.0. It adds 53 tokens to every session and 441 once invoked, about $0.0002 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 2026-10-17.
Other skills, from other repositories
notion
Notion workspace integration for searching pages, managing databases, creating postmortems, and exporting RCA findings.
jira
Jira integration for searching recent development context, tracking incidents, and managing issues during RCA.
slack
Slack integration tools for reading channel messages and thread replies.
incidentio
Skill "incidentio" from Arvo-AI/aurora, covering incident.io integration, instructions, tool usage (use in this order), investigation patterns and rca investigation workflow.
teams
Microsoft Teams integration for teammate messaging during incidents and @mention replies.
cloud-cost-agent
Use when analyzing cloud spend, running a cost audit, or finding savings across AWS, Google Cloud, and Azure — rightsizing, idle resources, and commitment/reservation recommendations, per-provider or cross-cloud. Uses only native CLIs and recommendation services (Cost Explorer/Compute Optimizer/Cost Optimization Hub…