docker-compose-manager
Fail
Audited by Gen Agent Trust Hub on Aug 31, 2026
Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [REMOTE_CODE_EXECUTION]: The Linux installation metadata specifies fetching and executing the official Docker installation script from get.docker.com.
- Evidence: 'linux: "curl -fsSL https://get.docker.com | sh && docker compose version"' in SKILL.md.
- [COMMAND_EXECUTION]: The skill performs shell commands using the docker compose CLI to manage containerized services.
- Evidence: Core capabilities in SKILL.md include commands such as 'docker compose up', 'docker compose ps', and 'docker compose down'.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external docker-compose.yml files which could contain instructions intended to influence the agent if the files are from untrusted sources.
- Ingestion points: 'compose_file' input parameter in SKILL.md.
- Boundary markers: None identified.
- Capability inventory: Shell execution of 'docker compose' commands as described in SKILL.md.
- Sanitization: No explicit validation or sanitization of the compose file content is specified.
Recommendations
- HIGH: Downloads and executes remote code from: https://get.docker.com - DO NOT USE without thorough review
Audit Metadata