Docker hardening

Skill eooo-io/ai-native-engineering-doctrine/04-agent-skills/docker-hardening

AI-native engineering harness: doctrine-only, fork-and-curate scaffold for standards, agent skills, tool adapters, and review rubrics.

Install

npx -y skills add eooo-io/ai-native-engineering-doctrine --skill docker-hardening

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Review Dockerfiles and Compose setups for security, reliability, image size, and developer ergonomics.

SKILL.md

1.5 KB, 280 tokens by cl100k_base, as published. Nobody here has run it

Docker Hardening

Purpose

Review a Dockerfile, Compose file, or build pipeline for the predictable failure modes: bloated images, leaked secrets, missing health checks, root-owned runtimes, and base images that haven't been updated since the previous coffee shortage.

Review areas

Base image choice and version pinning.
Dependency installation order and cache usage.
Layer caching effectiveness.
Non-root user and filesystem permissions.
Secret handling (environment, runtime injection, never bake-in).
Health checks and readiness signals.
Multi-stage builds and final image surface.
Reproducibility (locked dependencies, deterministic build inputs).
Image size and unused tooling left in the final stage.
Local developer usability (build time, dev/prod parity, debug story).

Output format

## Summary

## Security findings

## Reliability findings

## Image size and build findings

## Developer-ergonomics findings

## Recommended changes

Rules

Never bake secrets into images. Use environment variables, secret stores, or runtime injection.
Tie each finding to a concrete line or layer; "this image is too big" without a target is not a review.
Prefer minimal, reproducible base images over whatever the README first suggested.
Do not invent deployment-environment requirements; ask if they aren't stated.

Gives 0 of the 12 instructions most security skills give in 280 tokens

Counted across 648 of the 828 authors here whose files we hold, read 2026-08-06

parameterize all database queriesin 67 of 648, across 49 files
hash passwords using bcrypt scrypt or argon2in 48 of 648, across 35 files
apply rate limiting to authentication endpointsin 48 of 648, across 24 files
Configure security headersin 35 of 648, across 18 files
validate all inputsin 32 of 648, across 24 files
validate all external input at the system boundaryin 29 of 648, across 18 files
run containers as a non-root userin 28 of 648, across 15 files
use httponly secure samesite cookies for sessionsin 26 of 648, across 15 files
run dependency audits before every releasein 21 of 648, across 10 files
encode output to prevent cross-site scriptingin 21 of 648, across 10 files
copy dependencies before source codein 20 of 648, across 9 files
store secrets in environment variablesin 20 of 648, across 17 files

Said here and by no other author read

review base image choice and version pinning
evaluate dependency installation order and cache usage
assess layer caching effectiveness
verify non-root user and filesystem permissions
check secret handling
review health checks and readiness signals

Grouped from the skills themselves: near-identical wordings counted once, and counted by distinct author, so one author publishing three of these counts once. Length counted with cl100k_base; the agent that loads this file may tokenize it differently.