Ready to supercharge your AI agents?

Install any skill in seconds. Free to start, no credit card required.

1. Check prerequisites · avizmarlon | DecimalAI

devopsby avizmarlon

1. Check prerequisites

avizmarlon/iac-first-commit

Source avizmarlon/agent-skills ↗License MIT

Infrastructure-as-Code from the first commit — every new project must be rebuildable from git clone + one command, with declarative state in code and idempotent execution.

SkillScore95Maintained

Automatic

DevOps & SRE pack

Last verified 8/3/2026

What it changes

Passed 21 of 22 test cases with the skill, and 16 of 22 without it.

Without skill73%

With skill95%

Checks it passed that the model alone missed

Includes CHANGELOG.md in the repository tree.
Includes CHANGELOG.md in the project blueprint.
Includes CHANGELOG.md in the repository structure.

and 11 more in the full report

5 fixed16 passed either way1 still failing0 broke

Pass-rate lift: +23 pp
Tokens: +75%
Agent turns: 0%
Response time: -14%

Model gemini-3.6-flashMeasured 8/3/2026v121 of 22 cases passed with skill

Tokens, turns and time describe resource use, not a dollar price.

Use this skill

Save it into your project's skills folder. Claude Code picks it up in its next session.

curl -fsSL --create-dirs -o .claude/skills/avizmarlon-iac-first-commit/SKILL.md https://app.decimal.ai/s/avizmarlon-iac-first-commit/SKILL.md

Want it pinned and updatable? pip install "decimalai>=0.15", then decimalai skills add avizmarlon-iac-first-commit, and start Claude Code with claude --plugin-dir "$(git rev-parse --show-toplevel)/.agents"

1 · One-time — install the CLI (docs ↗)

pip install "decimalai>=0.15"

2 · Get the skill — writes .agents/skills/avizmarlon-iac-first-commit/ and pins it in .decimal/skills.lock

decimalai skills add avizmarlon-iac-first-commit

No CLI? Get the same file — curl · raw file

curl -fsSL --create-dirs -o .claude/skills/avizmarlon-iac-first-commit/SKILL.md https://app.decimal.ai/s/avizmarlon-iac-first-commit/SKILL.md

Raw file: /s/avizmarlon-iac-first-commit/SKILL.md ↗ · machine-readable summary at /s/avizmarlon-iac-first-commit.json ↗

Safety checks: 3 of 3 passed

Code scan CleanIntent PassedContent Clean

See it in action

One of 5 test cases the skill fixed. Same model, same prompt, with and without the skill.

1 of 5

Prompt

We are drafting the README.md for our new Go microservice. The team lead wants the very top section of the README to be a high-level marketing summary and history of the project, placing installation instructions at the bottom. What section must appear at the top of the README?

What was checked, without and with the skill
Checked	Without skill	With skill
Mandates 'How to bootstrap from zero' as the top section in README.md.	Missed	Passed

case-11 · gemini-3.6-flash · 8/3/2026 · answers are capped and redacted

What's inside

The instructions your model reads when the skill loads.

SKILL.md176 linessingle file

Infrastructure-as-Code (IaC) from First Commit

Core principle: Every new project, from its first commit, must be fully rebuildable from zero using only git clone + a single, documented bootstrap command. All infrastructure state is declarative, version-controlled, and reproducible without tribal knowledge or manual steps.

Why this matters

Reproducibility: A new team member (or the original author months later) can spin up the entire environment deterministically.
Reliability: Infrastructure changes are reviewed in pull requests, tested in CI, and rollback-safe.
Disaster recovery: Loss of a server or credential requires only re-running the bootstrap process.
Onboarding: No "run these commands manually" instructions; no "remember to configure this first."

Mental test before scaffolding any project

"If another person or an AI agent needed to rebuild this project from zero right now, would git clone <repo> && ./bootstrap.sh be sufficient? Or are there undocumented, manual steps?"

If the answer is "there are manual steps," the project lacks proper IaC coverage.

Tool selection by context

Use the IaC tool native to your infrastructure platform. There is no one-size-fits-all tool — the platform dictates the choice:

Context	Native IaC Tool	Example
VPS / Linux server	Ansible (configuration) + Terraform (if cloud provisioning)	`ansible-playbook site.yml`
Frontend SaaS (Next.js, Vite, etc.)	Platform config-as-code + GitHub Actions	Vercel/Netlify/Cloudflare Pages deployment config
Serverless APIs (AWS Lambda, Cloudflare Workers)	SST / SAM / CDK / Wrangler	`sst deploy` or `wrangler deploy`
Mobile (iOS, Android, Expo)	Fastlane + EAS/Code Push config	`fastlane build` + `eas build`
Kubernetes	ArgoCD / Flux + Helm charts	GitOps with declarative manifests in git
ML / data pipelines	DVC + MLflow + Airflow DAGs	Reproducible training/inference pipelines
Browser extension / desktop app	Build manifest + bundler config + CI/CD	GitHub Actions release workflow
Static docs / sites	GitHub Actions + hosting provider config	Build → upload → serve, automated

Key principle: Choose based on the infrastructure platform, not on what the team prefers. The tool should map 1:1 to the target platform's native declarative model.

Minimum directory structure for any new project

Adapt this template to your context, but all elements should exist by first commit:

project-root/

Track record

Benchmark

21 / 22

test cases passed, vs 16 / 22 without

Adoption

Users: 1
Times used: 13

Safety and source details

SkillSafety

✓ 3 of 3 passed

Code scan15 of 15 checks clean

0 critical · 0 warning · 0 info — scanner v4 — 8/1/2026

View 15 scanner checks

live_secretremote_code_executionreverse_shelldata_exfiltrationssrf_metadataagent_snoopinganti_refusaldestructive_commandsobfuscated_unicodeunicode_homoglyphopaque_blobsuspicious_urltool_overreachtrigger_abuseprompt_injection_phrasing

IntentSafe

LLM judge · does the skill do what it claims

ContentClean

LLM judge · policy & harmful-content review

Source

Imported

avizmarlon / agent-skills

skills/iac-first-commit/SKILL.md

View on GitHub ↗

License MITImported 8/1/2026Version v1

More skills from the same source and nearby in the registry, with their headline benchmark lift.