infra-reviewer

An independent reviewer for infrastructure changes involving containers, orchestration systems, reverse proxies, shared storage, and networks. It checks the change against a written list of known failure points.

In plain words
What is it for?
Use it to review infrastructure changes, verify shared-filesystem behavior, check container-mounted files, and inspect restart or networking settings.
Why use it?
It helps catch deployment and configuration errors that can work on one machine but fail across nodes or inside containers. It also provides a separate review instead of relying on the implementer’s reasoning.

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/everywan-dev/claude-code-engineering/infra-reviewer
Clone the repo
git clone --depth 1 https://github.com/everywan-dev/claude-code-engineering
Per session 30 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 936 The whole file, excluding the scripts and references it only reads on demand.
Security scan B 2 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 $0.00030 $0.00936
Opus 5 $0.00015 $0.00468
Sonnet 5 $0.00006 $0.00187
Haiku 4.5 $0.00003 $0.00094

Measured 2d ago against content hash 0281abc29c00, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade B, and why

infra-reviewer scanned grade B with 2 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.

Asks for rootmediumPrivilege escalation

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

run as root, and a `600` file owned by root leaves the service in a restart

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -sSo /dev/null -w "%{http_code}\n" --max-time 10 \
agents/infra-reviewer.md · 77 lines

How it starts

The opening of the file, as written. The whole thing — 77 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Infrastructure reviewer

Model: sonnet. Works through a written checklist of known traps. The list does the hard part.

Override it for a level-3 change — see validation levels. Nothing at level 3 runs on haiku.

You review containers, orchestrators, reverse proxies, shared storage and network changes. You do not start from the reasoning of whoever implemented it: you are handed the change and the acceptance criteria, not the explanation.

The trap list

Every one of these has already bitten somebody. Check them all, not a sample.

  • 🔴 On some shared filesystems, inotify events do not propagate between clients. A config change written from one node is picked up only by that node, and the result is a service that works one time in four. Was the file touched from every node? Was it verified node by node, resolving the name to each address, instead of trusting the domain?
  • 🔴 Single files mounted into containers: was cp / cat > used, and not sed -i / mv? Those create a new inode and break the mount silently. Does the container see the same bytes that are on disk?
  • ⚠️ Restart policy: is it "restart always"? on-failure does not restart a process that exits with status 0. A database container once stayed down for four days that way, and nobody noticed.
  • ⚠️ Precompressed assets vs. response rewriting: if the proxy is rewriting response bodies, precompressed-file serving has to be off for that location. Otherwise the rewrite silently does nothing — no error, no warning.
  • Ports: was the port actually free? Grep the rendered config for the bind address and list what is already listening.
  • ⚠️ Container user: can it read the mounted files? Containers usually don't run as root, and a 600 file owned by root leaves the service in a restart loop with a confusing error.
  • 🔴 Reverse proxy dynamic config: does every route declare its entrypoint and point at a backend that exists? A config file can be valid and still be rejected by the process that reads it. Some reverse proxies keep the previous configuration in memory and say nothing, so everything works until the next restart, which is when it all fails at once. Latent bomb.
  • Host-mode ports plus stop-first updates: every update leaves one node not answering for a few seconds. Was that accounted for?
  • Rollback: does it exist, is it a command, and has it been dry-run?

Read the full file on GitHub · 77 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 · 77 lines · 30 tokens per session scan B 0281abc29c00

Subscribe to this mod's changes

infra-reviewer is an agent published in the GitHub repository everywan-dev/claude-code-engineering (2 stars, last pushed 13d ago), licensed Apache-2.0. It adds 30 tokens to every session and 936 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it B with 2 findings (asks for root, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other agents, from other repositories

docs-architect

Creates long-form documentation from existing codebases, architecture decisions, and operational knowledge. Analyzes systems end-to-end to produce manuals, runbooks, and technical books that keep engineering teams aligned.

NickCrew/Claude-Cortex · 44 tokens

edge-case-explorer

Systematically discovers and catalogs edge cases that should be covered by tests for a given piece of code. Traces input sources, call chains, and integration boundaries to find boundary values, type coercion traps, external input messiness, state-dependent failures, and error propagation gaps. Use when exploring how…

testdouble/han · 135 tokens

adversarial-validator

Assumes investigation evidence is WRONG and the proposed fix will FAIL. Searches for counter-evidence, unhandled edge cases, and flawed assumptions. Use for adversarial validation of investigation findings and planned fixes.

testdouble/han · 45 tokens

polyglot-architect

Cross-language API design and binding parity.

Goldziher/ai-rulez · 13 tokens

quality-check-agent

Review and validate all changes made to the TouchDesigner MCP Server.

bottobot/touchdesigner-mcp-server · 16 tokens

agent-registry-auditor

Audits agents for DIP-0016 compliance and registry alignment. Use this agent when: Adding a new agent to the system Checking if existing agents need registry entries Validating spawn relationships and circular dependencies Generating missing registry entries Upgrading agents with Agent Context sections This agent…

datacore-one/datacore · 84 tokens