Back to skill
v1.0.0

Docker Essentials 1.0.0

BenignClawScan verdict for this skill. Analyzed May 1, 2026, 5:26 AM.

Analysis

This is an instruction-only Docker command reference; it includes powerful Docker operations that should be run deliberately, but the artifacts show no hidden code, credentials, or deceptive behavior.

GuidanceThis skill is reasonable as a Docker command reference. Before letting an agent run commands from it, explicitly confirm destructive cleanup commands, image pushes, volume mounts, port exposure, root exec sessions, and builds from unfamiliar Dockerfiles or images.

Findings (3)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

Abnormal behavior control

Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.

Tool Misuse and Exploitation
SeverityMediumConfidenceHighStatusNote
SKILL.md
docker rm -f container_name ... docker container prune ... docker-compose down -v ... docker system prune --volumes

These are destructive Docker cleanup/removal examples. They are expected for a Docker essentials reference, but they can delete containers, images, or volume-backed data if run without explicit intent.

User impactAn agent or user running these commands could remove development containers, images, or data volumes.
RecommendationRequire explicit user confirmation before running prune, remove, down -v, or other destructive Docker commands.
Unexpected Code Execution
SeverityMediumConfidenceHighStatusNote
SKILL.md
docker run -it ubuntu bash ... docker exec -u root -it container_name bash ... docker build --no-cache -t myapp .

Docker run, exec, and build commands execute containerized code or Dockerfile build steps. This is central to the skill purpose, but it depends on the trustworthiness of images, containers, and local build contexts.

User impactRunning untrusted images or Dockerfiles could execute unwanted commands inside containers and may interact with mounted host files or exposed ports.
RecommendationUse trusted images and Dockerfiles, review volume mounts and port mappings, and approve root exec/build actions before running them.
Agentic Supply Chain Vulnerabilities
SeverityInfoConfidenceHighStatusNote
SKILL.md
requires":{"bins":["docker"]} ... docker-compose up

The skill declares only the docker binary but includes docker-compose commands. This is a purpose-aligned dependency note, not evidence of malicious behavior.

User impactCompose workflows may fail or use an unexpected local Compose installation if docker-compose is not available or differs from Docker's compose plugin.
RecommendationEnsure Docker Compose is installed and use the intended command form for your environment, such as docker compose or docker-compose.