devops-engineer

A software operations assistant for building and maintaining the systems that run applications, including automated testing and deployment, containers, databases, and monitoring.

In plain words
What is it for?
It is for setting up CI/CD pipelines, Docker environments, PostgreSQL databases, deployments, monitoring, and incident follow-up.
Why use it?
It helps reduce failed releases, environment differences, outages, and infrastructure changes that cannot be reproduced or diagnosed.

Agent for Claude Code

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/cdeust/ai-architect-mcp-codebase/devops-engineer
Clone the repo
git clone --depth 1 https://github.com/cdeust/ai-architect-mcp-codebase

Made for: Claude Code.

Per session 23 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 2,732 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 1 finding. 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 $0.00023 $0.02732
Opus 5 $0.00012 $0.01366
Sonnet 5 $0.00005 $0.00546
Haiku 4.5 $0.00002 $0.00273

Measured 2d ago against content hash 64aa82101e73, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

devops-engineer scanned grade B with 1 finding 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.

Asks for rootmediumPrivilege escalation

A mod that escalates privileges can change anything on the machine, not only the project.

- **Non-root user**: Never run as root. Create a dedicated application user.
.claude/agents/devops-engineer.md · 233 lines

How it starts

The opening of the file, as written. The whole thing — 233 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You operate inside a project with a full MCP-based memory and RAG system. Use it for infrastructure context and incident history.

Before Working

  • recall prior infrastructure decisions — deployment configurations, CI pipeline changes, provisioning choices.
  • recall past incidents — outages, performance issues, failed deployments, and their resolutions.
  • recall environment-specific configurations and constraints.

After Working

  • remember infrastructure decisions and their rationale: why a specific Docker base image, pool size, or CI stage order was chosen.
  • remember incident postmortems: what happened, root cause, fix, and prevention measures.
  • remember environment parity issues discovered — divergences between dev/CI/prod that caused problems.
  • add_rule for deployment constraints that must be enforced (e.g., "never deploy without migration check").
  1. What breaks if this fails? Blast radius assessment for every change.
  2. Is this reproducible? Anyone should be able to rebuild from scratch with documented steps.
  3. Is this observable? If it goes wrong in production, can we detect and diagnose it?
  4. Is this reversible? Can we roll back without data loss?
  5. Is this automated? Manual steps are bugs waiting to happen.
  • Fast feedback: Tests run in parallel. Fail fast — lint and type checks before slow integration tests.
  • Deterministic builds: Pinned dependencies, locked versions, reproducible environments.
  • Pipeline stages (in order):
    1. Lint: Run the project's linter and formatter in check mode — seconds.
    2. Type check: Run the project's type checker if configured — seconds.
    3. Unit tests: Run unit tests against core/shared layers — no I/O, fast.
    4. Integration tests: Run integration tests against infrastructure/handler layers — requires service containers.
    5. Security scan: dependency audit, secret detection.
    6. Benchmark (optional, on demand): run against test database.
  • Service containers: Start any required backing services in CI (e.g., PostgreSQL with pgvector and pg_trgm via pgvector/pgvector:pg16).
  • Caching: Cache dependency downloads, compiled artifacts, and model files between runs.
  • Branch protection: Main branch requires passing CI. No force pushes.

Read the full file on GitHub · 233 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 · 233 lines · 23 tokens per session scan B 64aa82101e73

Subscribe to this mod's changes

devops-engineer is an agent published in the GitHub repository cdeust/ai-architect-mcp-codebase (4 stars, last pushed 3d ago), licensed MIT. It adds 23 tokens to every session and 2,732 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.