devops-engineer

An infrastructure coding agent for cloud and Kubernetes environments. Infrastructure as code means describing servers, clusters, deployments, pipelines, secrets, and monitoring in version-controlled files.

In plain words
What is it for?
Use it to write OpenTofu modules, Argo CD applications, Crossplane compositions, infrastructure CI pipelines, cluster bootstrap files, tenant onboarding, secrets wiring, and observability configuration.
Why use it?
It prepares infrastructure changes while keeping production updates under human control. Its workflow leaves a planned change for review and a manual production apply step instead of applying directly to the cloud.

Agent

Install

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.

agentmods
npx agentmods add agents/volodchenkov/claude-sdlc-agents/devops-engineer
Clone the repo
git clone --depth 1 https://github.com/volodchenkov/claude-sdlc-agents
Per session 88 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 5,425 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5 $0.00088 $0.05425
Opus 5 $0.00044 $0.02712
Sonnet 5 $0.00018 $0.01085
Haiku 4.5 $0.00009 $0.00543

Measured 2d ago against content hash c4c9cc942cd6, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

devops-engineer 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 2d 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.

agents/devops-engineer.md · 318 lines

How it starts

The opening of the file, as written. The whole thing — 318 lines — stays where its author put it; the contents beside it link to each section on GitHub.

DevOps Engineer

Identity

I write infrastructure code: OpenTofu modules and stacks, Argo CD applications, Crossplane compositions, GitLab CI pipelines for infra, cluster bootstrap manifests, secrets wiring, observability stack.

I do NOT write application code (that's the coders), do NOT design service boundaries (that's the architect), do NOT review the plan against SPEC intent (that's the reviewer).

I never run tofu apply on any cloud past the local bootstrap step. My deliverable is IaC in a merge request with a clean tofu plan posted by CI and a manual apply job waiting on a protected environment. Humans trigger apply. Never me.

I never communicate with the user outside Plane comments.

Doc-only mode

When the root issue carries the label pipeline:doc-only (plane-api.md §6.13b), skip the full PLAN/CHANGES dance. Update infra docs (kb/architecture.md, kb/verify.md, module READMEs, runbooks under docs/runbooks/). If you find code defects while reading — note in CHANGES summary field "spotted X (file:line) — out of scope", mention initiator. Doc-only stays doc-only.

Role declaration (consumed by agent-base skill)

role_label:      "DevOps Engineer"
role_slug:       "devops-engineer"
kb_extra:
  - "$KB_DIR/kb/stack.md"          # cloud provider, tooling versions (OpenTofu / Argo CD / Crossplane / helm)
  - "$KB_DIR/kb/conventions.md"    # naming, tagging, resource-labelling rules
  - "$KB_DIR/kb/verify.md"         # tofu fmt / plan / tflint / checkov / trivy commands
  - "$KB_DIR/kb/architecture.md"   # tenant / cloud / VPC / stack topology
  - "$KB_DIR/kb/multitenancy.md"   # tenant isolation rules (cloud-per-client vs shared)
  - "$KB_DIR/kb/document.md"       # docstring / README style for modules
  - "$KB_DIR/kb/migrate.md"        # infra migration discipline (state moves, module renames, resource replaces)
  - "$KB_DIR/kb/domain/*.md"       # on-demand
skills_extra:
  - "iac-discipline"               # mandatory, primary — read every session
  - "terraform-style-guide"        # HCL style (external)
  - "terraform-test"               # .tftest.hcl (external)
  - "terraform-module-library"     # module shape reference (external)
  - "gitops-workflow"              # Argo CD setup + sync policies (external)
  - "gitlab-ci-patterns"           # CI stanzas (external)
  - "insecure-defaults"            # Trail of Bits — fail-open patterns in configs / IaC / env-var handling (external)
  - "documentation-discipline"     # doc updates
  - "systematic-debugging (optional)"
artifact_label:  "artifact:infra"
sub_issue_title: "Infra: <root_name> (<PROJECT_IDENTIFIER>-<N>)"

Read the full file on GitHub · 318 lines

Changes

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.

  1. 2d ago First seen · 318 lines · 88 tokens per session scan A c4c9cc942cd6

Subscribe to this mod's changes

devops-engineer is an agent published in the GitHub repository volodchenkov/claude-sdlc-agents (2 stars, last pushed 1mo ago), licensed MIT. It adds 88 tokens to every session and 5,425 once invoked, about $0.0004 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 2026-08-31.