Atmos is an infrastructure runtime that coordinates tools such as Terraform, OpenTofu, Kubernetes, Helm, Packer, Ansible, and containers through consistent commands and configuration. It is for teams running cloud infrastructure on laptops, in CI, or through AI agents across environments and regions. Its catalogue entries provide skills, agents, commands, and other add-ons for Atmos workflows.
Borrowing it
Nothing to install: this file belongs to cloudposse/atmos. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/cloudposse/atmos/main/.claude/skills/code-hygiene/SKILL.mdgit clone --depth 1 https://github.com/cloudposse/atmosWrote 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/cloudposse/atmos/code-hygiene)<a href="https://agentmods.dev/skills/cloudposse/atmos/code-hygiene"><img src="https://agentmods.dev/badge/skills/cloudposse/atmos/code-hygiene/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/cloudposse/atmos/code-hygiene"><img src="https://agentmods.dev/badge/skills/cloudposse/atmos/code-hygiene.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00225 | $0.02777 |
| Opus 5.5 | $0.00090 | $0.01111 |
| Sonnet 5.5 | $0.00045 | $0.00555 |
| Haiku 4.5 | $0.00022 | $0.00278 |
Grade A, and why
code-hygiene 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 — 167 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Code Hygiene
A narrow, calibrated review pass — not a general bug hunt. lint catches syntactic violations,
test-coverage catches untested lines, and a normal correctness-focused review catches bugs — none
of those structurally catch an abstraction that got duplicated instead of reused, a feature that
validates but silently does nothing, or a comment that admits a gap nobody closed. This skill's job
is specifically the class of smell that only shows up when you read code against this repo's own
architectural mandates (CLAUDE.md) and ask "does this actually do what it claims to," not just
"does this compile and pass its own tests."
This skill reports; it does not redesign. A real architectural smell usually needs a design decision, not a mechanical patch — report it clearly enough that a human (or a follow-up planning pass) can make that call. See "Auto-fix policy" below for the one narrow exception.
Scope: patch-aware by default
Default to the same scoping convention as lint/test-coverage: review only the files/packages
touched by git diff origin/main...HEAD --name-only (or the currently open PR's diff). This never
goes hunting through the other 340+ packages the current patch didn't touch. Unlike lint (which
wraps a real atmos fix lint command with an actual --new-from-rev flag), this skill has no
underlying CLI command — there's no literal flag to pass, only the phrasing below.
Full-repo mode (explicit only)
Only when a human explicitly asks for a full sweep (e.g. "audit the whole repo", "full code
hygiene sweep" — never inferred, never run from fix-all's automated cycle). A whole-repo pass is
expensive and belongs in an on-demand invocation, not an hourly loop.
Dedup: skip a re-run against an unchanged diff
Before reviewing, hash git diff origin/main...HEAD (the diff content itself — this skill's
dedup key is "has the patch changed," not an external event set) and compare against
.claude/state/code-hygiene/<branch-slug>.json's last-recorded hash. The .claude/state/<name>/ <branch-slug>.json file shape is borrowed from .claude/hooks/security-remediate-trigger.sh — the
one existing precedent for this pattern in this repo, not an established multi-skill convention —
adapted here to hash the diff instead of a set of alert IDs, since what's being deduped against is
different (a review cycle re-running on unchanged code, not a repeated external notification). If
the hash matches, skip re-reviewing and reuse the cached findings array
from that state file — report it as a one-line no-op citing the cached result count, don't burn
tokens re-deriving the same answer from an unchanged patch. If the hash differs (or no state file
exists yet), do the full review below, then write {hash, findings, ts} back to the state file
after reporting.
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 · 167 lines · 225 tokens per session scan A 113e5c038ff6
code-hygiene is a skill published in the GitHub repository cloudposse/atmos (1,403 stars, last pushed today), licensed Apache-2.0. It adds 225 tokens to every session and 2,777 once invoked, about $0.0009 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 2026-08-30.
Other skills, from other repositories
apply-qa-fixes
Apply QA gate findings then hand back for re-review. Never self-approves or closes. Use when: apply QA fixes, remediate gate FAIL/CONCERNS, /apply-qa-fixes.
aiox-qa
Test Architect & Quality Advisor (Quinn). Use for comprehensive test architecture review, quality gate decisions, and code improvement. Provides thorough analysis including requ...
hatch3r-refactor
Internal code quality improvement workflow without changing external behavior. Use when refactoring code structure, simplifying modules, or improving maintainability.
looper-qa
Use when a Looper-managed GitHub repo needs scheduled pre-merge QA — a PR carries the looper:qa label, the spec stage reaches looper:spec-ready, or the Looper reviewer loop requests an independent second pass. Runs the full QA cycle (Looper state probe → PR checkout → ffs code review → language-specific test suite →…
remove-ai-slops
Removes AI-generated code smells from branch changes or an explicit file list behind regression tests. Use when the user asks to clean up, deslop, or remove AI-slop patterns from recent changes.
review-work
Post-implementation gate review: run manual QA on the real surface yourself, then launch ONE gate reviewer (never a panel) to audit goal, constraints, code quality, security, missed context, and QA evidence. Use before a PR handoff or when the user explicitly asks to review completed work.