agentleFS
abevzabevz/devops-skills/skills/gitops-review/SKILL.md

gitops-review

Skill0 starsChanged 41 days agoMITView on GitHub

Use when reviewing GitOps repository structure, environment separation, ownership, or promotion flow, independent of the controller. Mention "gitops review", "deploy repo structure", or "gitops repo layout"; use argocd for controller operation, argocd-debug for one stuck Application, and argocd-applicationset for generator design.

1 other project uses a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

6 rules. About 0 are common, 5 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈5

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the repository tree, environment boundaries, application definitions, and pre-merge checks.

    line 18
  2. Distinctive:

    Do not mistake a single manifest finding for a repository-flow finding.

    line 22
  3. Distinctive:

    Follow the Workflow, Safety rules, and Quality checklist below.

    line 26
  4. Distinctive:

    Use when reviewing how a GitOps repository is organized and how changes flow from commit to running cluster — layout, environment …

    line 37
  5. Very specific:

    Use argocd instead for controller operation, argocd-debug for one stuck Application, and argocd-applicationset for generator/template …

    line 39
  6. Distinctive:

    Treat "no automated policy checks pre-merge" as a real risk to call out, not just a nice-to-have.

    line 84

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

GitOps Review

TL;DR checklist

  • Trace where an Application is defined and how a commit reaches each environment.
  • Check secrets, image updates, drift handling, rollback, and bootstrap.
  • Report structural gaps with the affected path and promotion risk.

Key read-only checks

  • Read the repository tree, environment boundaries, application definitions, and pre-merge checks.

Common pitfalls

  • Do not mistake a single manifest finding for a repository-flow finding.

Agent procedure

Follow the Workflow, Safety rules, and Quality checklist below.

Quick references

  • references/validation-policy.md
  • references/tools.md

Last verified: unverified

When to use

Use when reviewing how a GitOps repository is organized and how changes flow from commit to running cluster — layout, environment separation, secrets, and promotion process. Use argocd instead for controller operation, argocd-debug for one stuck Application, and argocd-applicationset for generator/template design.

Goal

Find structural gaps in the GitOps repo/process that cause drift, unsafe promotions, or unclear ownership — not a review of any single manifest.

Workflow

  1. Layout — is the repo organized as app-of-apps, ApplicationSet-driven, or a flat list of Applications? Is the pattern consistent, or mixed in ways that make it unclear where a given app is defined?
  2. Environment separation — are dev/staging/prod separated by directory, branch, or repo? Is it possible to accidentally promote a change to prod by editing the wrong directory, or does the structure prevent that class of mistake?
  3. Secrets strategy — how are secrets handled (Sealed Secrets, External Secrets Operator, SOPS, Vault)? Confirm no plaintext secrets are committed, and that the secrets tool used is consistent across the repo.
  4. Image update flow — how do new image versions get into the repo (manual PR, image updater/automation)? Is there a review gate before deploy, or does an image push go straight to prod?
  5. Drift handling — is there automated drift detection/self-heal, and is it safe for every environment (e.g. selfHeal: true might be desired in dev but risky if it silently reverts an intentional emergency kubectl change in prod without alerting anyone)?
  6. Rollback strategy — is rollback "revert the Git commit" (fast, auditable) or does it require manual cluster intervention? Is the revert path documented?
  7. Policy checks — are manifests validated pre-merge (schema validation, OPA/Kyverno policy checks, conftest) or only after they've already synced to a cluster?
  8. Promotion process — how does a change move dev → staging → prod? Is it a PR/merge between directories or branches, and is there a required approval step before prod?
  9. Bootstrap safety — how is a new cluster/environment bootstrapped into this GitOps setup? Is the bootstrap process itself version-controlled and repeatable, or manual/tribal knowledge?

References

  • references/validation-policy.md — validation pipeline order (schema → lint → policy → cluster dry-run), Kyverno/Gatekeeper policy examples, and Kustomize overlay structure patterns and pitfalls (patch ordering, base drift, components).
  • references/tools.md — optional local tools (kustomize build, conftest, yamllint, gitleaks) that can supplement this manual review. Never install these automatically.

Safety rules

  • This is a review skill: do not modify the GitOps repo's deployed state or trigger a sync.
  • Treat "no automated policy checks pre-merge" as a real risk to call out, not just a nice-to-have.
  • Flag any path where a single commit can silently reach production with no review gate.

Output format

## Repo layout assessment
## Environment separation issues
## Secrets strategy issues
## Image update / promotion flow issues
## Drift handling / rollback issues
## Recommended changes, prioritized

Quality checklist

  • Environment-separation mistakes (accidental prod promotion) were explicitly considered
  • Secrets strategy was verified, not assumed safe
  • Rollback path (git revert vs. manual) was checked
  • Pre-merge policy validation presence was checked
  • No sync or deploy action was taken

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.