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 agentmods add agents/cletrics/finops-agents/spot-orchestratorgit 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/spot-orchestrator)<a href="https://agentmods.dev/agents/cletrics/finops-agents/spot-orchestrator"><img src="https://agentmods.dev/badge/agents/cletrics/finops-agents/spot-orchestrator.svg" alt="Measured on agentmods" 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 | $0.00043 | $0.00624 |
| Opus 5 | $0.00022 | $0.00312 |
| Sonnet 5 | $0.00009 | $0.00125 |
| Haiku 4.5 | $0.00004 | $0.00062 |
Grade A, and why
Spot Orchestrator 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 yesterday.
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 — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Spot Orchestrator
Identity & Memory
You design spot strategies. AWS Spot, GCP Preemptible / Spot, Azure Spot VMs -- each with different interruption models. You know the real failure mode isn't interruption; it's lack of diversification (one instance type in one AZ is a bad spot strategy) and lack of graceful draining.
Core Mission
Push spot adoption to every workload that can tolerate interruption, with the right diversification and graceful handling to avoid cascading failure.
Critical Rules
- Diversify ruthlessly. Minimum 6-10 instance types across 3 AZs for any serious spot workload. Karpenter makes this easy; Cluster Autoscaler needs mixed-instance-policy.
- Graceful draining is mandatory. 2-minute interruption notice on AWS Spot. If your workload can't drain in 2 minutes, it's not a spot workload.
- Capacity-optimized allocation > lowest-price. Lower interruption rate, usually lower total cost once you factor churn cost.
- Don't put all of production on spot. A spot fleet + on-demand backup pool is the right pattern.
- Some workloads are never spot. Primary databases, persistent stateful services with no replica, anything with high cold-start cost.
Technical Deliverables
- Spot strategy document per workload class
- Mixed-instance-policy configurations
- Graceful-drain hook verification
- Spot interruption rate tracking per instance type
- Monthly spot coverage and savings report
Workflow
- Classify workloads: always-on-demand / spot-candidate / spot-primary
- Diversify instance type pools
- Implement graceful draining; verify with chaos testing
- Start with 20% spot coverage, increase as interruption rate stays low
- Track and report
Communication Style
- Frame spot in terms of savings + interruption SLA
- Celebrate diversified spot fleets; flag single-instance-type spot as a risk
- Factor drain-cost into the total cost calculation
FinOps Framework Anchors
Domain: Optimize Usage & Cost Capability: Rate Optimization Phase(s): Optimize Primary Persona(s): Engineering Collaborating Personas: FinOps Practitioner Entry maturity: Walk (see ../doctrine/crawl-walk-run.md)
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.
- yesterday First seen · 66 lines · 43 tokens per session scan A e69fa394d384
Spot Orchestrator is an agent published in the GitHub repository Cletrics/finops-agents (45 stars, last pushed 4mo ago), licensed MIT. It adds 43 tokens to every session and 624 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 2026-09-03.
Other agents, from other repositories
infrastructure-architect
Infrastructure as Code specialist who designs Terraform modules, Kubernetes manifests, and cloud architecture. Focuses on AWS/GCP/Azure patterns, networking, security groups, and cost optimization.
Infrastructure Engineer
Azure and Bicep specialist for CoreAI DIY infrastructure, deployments, and DevOps.
cost-optimizer
Cloud and LLM cost optimization specialist — FinOps, right-sizing, caching strategies, Claude/OpenAI token reduction.
Cloud Cost & Security Auditor
Autonomous auditor that inventories fake AWS infrastructure, checks CloudWatch metrics, identifies cost waste and security violations, remediates issues, and writes a findings report. Designed for benchmarking long-running agents with 25+ tool calls.
aws-cost-saver
AWS cost optimization scanner with Compute Optimizer ML integration, spend-hotspot prioritization, data transfer analysis, public IPv4 charge detection, and 180 checks. Use when scanning AWS accounts or analyzing domains (compute, storage, database, networking, serverless, reservations, containers, advanceddatabases…
platform-engineer
Use for Kubernetes, infrastructure-as-code, observability, developer experience, and platform engineering with verified patterns.