Agent skill

Spotinfo

by alexei-led in alexei-led/spotinfo

Query Spot/preemptible VM prices, savings and interruption risk across AWS, GCP and Azure with the spotinfo CLI.

Apache-2.0Auto-check passedDevOps & Cloud

Install Spotinfo

skills CLI
$ npx skills add alexei-led/spotinfo --skill spotinfo -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install alexei-led/spotinfo spotinfo --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/alexei-led/spotinfo.git skills-src && mkdir -p .claude/skills && cp -r skills-src/skills/spotinfo .claude/skills/spotinfo && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
spotinfo
GitHub stars
164
Token cost
~1.8k tokens
SKILL.md length
706 words
Files
5 (incl. scripts, references)
Skills in repo
1
Repo updated
First seen
Licence
Apache-2.0

At a glance

Query Spot/preemptible VM prices, savings and interruption risk across AWS, GCP and Azure with the spotinfo CLI.

  • Asks which instance type is cheapest
  • SKILL.md covers Two habits, Three things that will break a…, unavailable is not zero and Flags that matter, plus 4 more sections
  • Runs Shell scripts from its folder
  • What a Spot instance costs

What it does

Spotinfo is an agent skill from alexei-led/spotinfo. Query Spot/preemptible VM prices, savings and interruption risk across AWS, GCP and Azure with the spotinfo CLI. Use this whenever the user asks which instance type is cheapest, what a Spot instance costs, what the interruption risk is, whether arm64 beats x8664 on price, how AWS/GCP/Azure Spot pricing compares, or wants to size a workload to a budget — including when they describe the need without naming spotinfo ("find me a cheap 8-core box", "least interruption-prone instance for a CI runner", "compare spot…

Its SKILL.md is about 1.8k tokens, which your agent loads only when the skill is triggered. The skill folder holds 6 other files, including scripts and reference files (for example `references/clouds.md`, `references/output-schema.md` and `references/recipes.md`). Compatibility notes: Requires the spotinfo binary on PATH (brew install alexei-led/tap/spotinfo, or ghcr.io/alexei-led/spotinfo). jq is needed only for scripts/compare-clouds.sh…

It sits in DevOps & Cloud, covering Landing pages, Web scraping and Cloud cost optimization. It works with Google Cloud, Amazon Web Services, Microsoft Azure and Model Context Protocol. The repository describes itself as: CLI for exploring Spot instance pricing across AWS, GCP and Azure. Inspect Spot instance types, savings, price, and AWS interruption frequency — offline, from embedded data. The licence is Apache-2.0.

When your agent uses it

  • Asks which instance type is cheapest
  • What a Spot instance costs
  • What the interruption risk is
  • Whether arm64 beats x8664 on price

Example prompts

  • “find me a cheap 8-core box”
  • “least interruption-prone instance for a CI runner”
  • “compare spot prices across clouds”
  • “/spotinfo”

Requirements

  • A Bash shell
  • Compatibility (from SKILL.md): Requires the spotinfo binary on PATH (brew install alexei-led/tap/spotinfo, or ghcr.io/alexei-led/spotinfo). jq is needed only for scripts/compare-clouds.sh. No cloud credentials and no network are required.

What it can do on your machine

Read from SKILL.md and the folder at commit a1306dc. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Ships 1 file in scripts/ (Shell), which the agent can run.

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

  • Compatibility

    Requires the spotinfo binary on PATH (brew install alexei-led/tap/spotinfo, or ghcr.io/alexei-led/spotinfo). jq is needed only for scripts/compare-clouds.sh. No cloud credentials and no network are required.

    From compatibility in the SKILL.md frontmatter.

Context cost

Spotinfo loads about 1.8k tokens when it runs, and up to ~6.3k if it reads all its reference files. Until then it costs about 178 tokens; SKILL.md has 706 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~178
When it runs · the whole SKILL.md, loaded when a task matches
~1.8k
With references · SKILL.md plus every file in references/, read only if the agent opens them
~6.3k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); the scripts in this folder are not scanned.

SKILL.md

The full file from alexei-led/spotinfo at commit a1306dc, republished under its Apache-2.0 licence (© alexei-led). 706 words, ~1,848 tokens.

Download SKILL.mdSave it as .claude/skills/spotinfo/SKILL.md (or your agent's skills folder). This skill also uses 4 other files; get the full folder from GitHub.
name
spotinfo
description
Query Spot/preemptible VM prices, savings and interruption risk across AWS, GCP and Azure with the spotinfo CLI. Use this whenever the user asks which instance type is cheapest, what a Spot instance costs, what the interruption risk is, whether arm64 beats x86_64 on price, how AWS/GCP/Azure Spot pricing compares, or wants to size a workload to a budget — including when they describe the need without naming spotinfo ("find me a cheap 8-core box", "least interruption-prone instance for a CI runner", "compare spot prices across clouds"). It answers offline with no cloud credentials in about 40ms, so prefer it over guessing prices from memory, scraping a pricing page, or writing throwaway SDK code.
compatibility
Requires the spotinfo binary on PATH (brew install alexei-led/tap/spotinfo, or ghcr.io/alexei-led/spotinfo). jq is needed only for scripts/compare-clouds.sh. No cloud credentials and no network are required.
license
Apache-2.0
metadata.author
alexei-led
metadata.version
3.0.0
metadata.repository
https://github.com/alexei-led/spotinfo

spotinfo

One binary that answers two questions about Spot/preemptible VMs on AWS, GCP and Azure:

  • "What is there?" → spotinfo list — requires nothing, shows every match
  • "What should I run?" → spotinfo recommend — requires --architecture, --min-vcpu and --min-memory-gib, returns a ranked page

Pick recommend when the user states a requirement ("I need 8 cores and 32 GB"). Pick list when they want to see what exists ("what do m5 instances cost").

It ships an embedded price snapshot per cloud, so it answers with no credentials and no network. A query is free and cannot fail on missing auth — run it instead of guessing a price from memory.

Two habits

Pass --offline unless the user needs today's price. It answers from the snapshot in about 40 ms and makes no network call. Without it, --region all (the default) queries every AWS region and takes seconds.

Pass --output json whenever you will parse the result. table is for people to read.

bash
spotinfo recommend --cloud aws --offline --output json \
  --architecture x86_64 --min-vcpu 8 --min-memory-gib 32 --top 5

Three things that will break a parser

The row key differs by command. list puts rows under candidates, recommend under recommendations. The documents otherwise look alike, so a parser written for one silently finds nothing in the other.

Prices are strings. "0.047500000" is fixed-point on purpose, because a float drifts on the last digit. Call tonumber before you compare or total. On list the field is nullable — AWS omits some machines from its price feed, and those carry null rather than 0.

risk is always an object, even when empty. Read risk.status first.

Full field reference: references/output-schema.md.

unavailable is not zero

On AWS, risk carries a real Spot Advisor measurement. On GCP and Azure every field except status is null:

json
"risk": { "status": "unavailable", "kind": null, "label": null }

That means the vendor does not publish the figure. It does not mean the risk is low, and it does not mean zero. Reading risk.label unconditionally gives you null, which prints as an empty cell and reads to a person as "fine".

When you report a GCP or Azure machine, say the cloud publishes no interruption figure. Quietly dropping the column is how someone puts a database on a preemptible VM.

This is also why the risk-capped workloads are AWS-only — the web, ci and batch ceilings are AWS Spot Advisor bucket boundaries:

bash
spotinfo recommend --cloud aws --workload web  --offline --architecture x86_64 --min-vcpu 4 --min-memory-gib 16
spotinfo recommend --cloud gcp --workload cost --offline --architecture x86_64 --min-vcpu 4 --min-memory-gib 16

--workload cost is the default, makes no interruption claim, and answers everywhere.

Per-cloud capabilities and limits: references/clouds.md.

Show full SKILL.md (320 more words)Show less

Flags that matter

FlagNote
--cloud aws|gcp|azureDefaults to aws. Set it whenever the user named a cloud.
--offlineSnapshot only, no network. Use by default.
--machineAn RE2 regexp, not a glob. '^m5\.', never 'm5*'. Anchor it or m5dn matches too.
--regionRepeatable. Defaults to all. Scope it when the user named a region.
--sort / --ordermachine|price|region|risk|savings|score, asc|desc.
--max-priceUSD per machine-hour ceiling.
--os linux|windowsAWS and Azure serve both. GCP refuses Windows.
--toprecommend only, maximum 50.

When it refuses

spotinfo refuses a flag it cannot honour instead of ignoring it. Exit is non-zero with empty stdout, so branch on the exit code and read stderr — the message names the flag, the cloud and the limit, and usually names the fix.

You ranIt says
--os windows --cloud gcpGoogle publishes Linux Spot prices only
--workload web on GCP/Azureno comparable interruption figure — use --workload cost
--sort score without --with-scoreonly --with-score fetches the figures the sort orders
--with-score --offlineno snapshot carries a placement figure
--cpu, --memory, --type, --priceretired names — the hint gives the replacement

Recipes

Common tasks with commands and result handling: references/recipes.md.

To compare one requirement across all three clouds in a single table:

bash
scripts/compare-clouds.sh --min-vcpu 4 --min-memory-gib 16 --offline

Machine naming differs per cloud, so there is no single cross-cloud query — the script runs one recommend per cloud and merges the output.

MCP server

spotinfo --mcp speaks MCP 2025-11-25 over stdio with three read-only tools: list_spot_machines, recommend_spot_machines, list_cloud_regions. Argument names mirror the flags mechanically (--min-vcpu → min_vcpu, --region → regions, an array).

If you already have shell access, prefer the CLI — one call, same schema, and you can pipe it.

Reporting to the user

Give price, savings percent and risk together. A cheap machine with a >20% interruption rate is a different recommendation from a cheap stable one, and the price alone hides that. When risk is unavailable, say the cloud does not publish it — never let silence read as safety.

© alexei-led, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 4 other files (scripts, references) in skills/spotinfo of alexei-led/spotinfo.

  • SKILL.md
  • references/clouds.md
  • references/output-schema.md
  • references/recipes.md
  • scripts/compare-clouds.sh

Open the folder on GitHubat commit a1306dc

Compare with similar skills

Spotinfo next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Spotinfo compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Spotinfo this skillalexei-led/spotinfo164—~1.8kAutomated safety check: PassApache-2.0
Cloud Cost Optimizationwshobson/agents40k14 repos~1.7kAutomated safety check: PassMIT
Terravision Cloud Diagramspatrickchugh/terravision1.6k—~5.6kAutomated safety check: NotesAGPL-3.0-only
FrugalyuanboP/frugal198—~2.1kAutomated safety check: PassMIT
Cloud ArchitectJeffallan/claude-skills12k—~1.9kAutomated safety check: PassMIT
Cloud Architectdavila7/claude-code-templates33k8 repos~1.9kAutomated safety check: PassMIT

Similar skills

  • Cuts cloud spend across AWS, Azure, GCP and OCI with cost tagging, rightsizing, commitment and spot pricing models, and architecture changes.

    40k GitHub starsUsed in 14 repos~1.7k tokens
    DevOps & CloudAuto-check passed
  • Terravision Cloud Diagrams

    patrickchugh/terravision

    Draw cloud architecture diagrams for AWS, Azure or GCP with the official provider icon sets, using TerraVision.

    1.6k GitHub stars~5.6k tokensUpdated 3 days ago
    DevOps & CloudAuto-check: notes
  • Frugal

    yuanboP/frugal

    Cloud cost awareness for agents. An agent skill from yuanboP/frugal.

    198 GitHub stars~2.1k tokensUpdated 2 mo ago
    DevOps & CloudAuto-check passed
  • Cloud Architect

    Jeffallan/claude-skills

    Designs cloud architectures, migration plans, cost optimization recommendations and disaster recovery strategies across AWS, Azure and GCP.

    12k GitHub stars~1.9k tokensUpdated 6 days ago
    DevOps & CloudAuto-check passed
  • Cloud Architect

    davila7/claude-code-templates

    Expert cloud architect specializing in AWS/Azure/GCP multi-cloud infrastructure design, advanced IaC (Terraform/OpenTofu/CDK), FinOps cost optimization, and modern architectural patterns.

    33k GitHub starsUsed in 8 repos~1.9k tokens
    DevOps & CloudAuto-check passed
  • Resource Tagging

    ancoleman/ai-design-components

    Apply and enforce cloud resource tagging strategies across AWS, Azure, GCP, and Kubernetes for cost allocation, ownership tracking, compliance, and automation.

    525 GitHub stars~4k tokensUpdated 10 mo ago
    DevOps & CloudAuto-check passed

Categories

Questions about Spotinfo

What does Spotinfo do?

Query Spot/preemptible VM prices, savings and interruption risk across AWS, GCP and Azure with the spotinfo CLI. Spotinfo is an agent skill from alexei-led/spotinfo. Query Spot/preemptible VM prices, savings and interruption risk across AWS, GCP and Azure with the spotinfo CLI.

When should I use Spotinfo?

Spotinfo fits situations like: asks which instance type is cheapest; what a Spot instance costs; what the interruption risk is; whether arm64 beats x8664 on price.

How do I install Spotinfo in Claude Code?

Run `npx skills add alexei-led/spotinfo --skill spotinfo -a claude-code`. Or copy the skill folder (skills/spotinfo in alexei-led/spotinfo) into .claude/skills/spotinfo in your project. Claude Code loads it when a task matches its description.

How do I install Spotinfo in Codex?

Run `npx skills add alexei-led/spotinfo --skill spotinfo -a codex`. Or copy the skill folder (skills/spotinfo in alexei-led/spotinfo) into .agents/skills/spotinfo in your project. Codex loads it when a task matches its description.

Can I use Spotinfo in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add alexei-led/spotinfo --skill spotinfo -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/spotinfo, .gemini/skills/spotinfo, .github/skills/spotinfo and .opencode/skills/spotinfo in your project.

What does Spotinfo need to run?

Going by SKILL.md and its folder, Spotinfo needs a shell for the scripts in its folder. Our summary lists: A Bash shell. Compatibility (from SKILL.md): Requires the spotinfo binary on PATH (brew install alexei-led/tap/spotinfo, or ghcr.io/alexei-led/spotinfo). jq is needed only for scripts/compare-clouds.sh. No cloud credentials and no network are required..

Does Spotinfo access the network?

SKILL.md contains no URLs. Any network use would come from the scripts or tools the agent runs. This is read from the text; nothing was executed.

Is Spotinfo safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. The check reads SKILL.md only: the scripts in the folder are not scanned, so read them before running anything.

What licence does Spotinfo use?

Spotinfo is published under the Apache-2.0 licence (declared in SKILL.md). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Spotinfo use?

About 1.8k tokens (SKILL.md is roughly 7.4k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full. Its references folder adds about 4.4k tokens, read only when the agent opens those files.

What are the alternatives to Spotinfo?

Skills that share tags, products or a category with Spotinfo: Cloud Cost Optimization (wshobson/agents, 40k stars), Terravision Cloud Diagrams (patrickchugh/terravision, 1.6k stars), Frugal (yuanboP/frugal, 198 stars) and Cloud Architect (Jeffallan/claude-skills, 12k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Spotinfo?

alexei-led (a GitHub user) maintains it in alexei-led/spotinfo, which has 164 GitHub stars. The repository was last updated on October 5, 2026.

Source: alexei-led/spotinfo on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.