scbe-docker-workflow
Pass
Audited by Gen Agent Trust Hub on Jun 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill provides instructions to execute several local commands including 'docker', 'docker-compose', and 'npm' scripts for building, running, and managing containerized applications.
- [COMMAND_EXECUTION]: Instructs the agent to execute a local PowerShell script 'scripts/scbe_docker_status.ps1' to perform repository diagnostics and status checks.
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface: The skill reads and processes configuration files from the user's repository which could contain malicious content designed to influence agent behavior. Ingestion points: 'Dockerfile', 'docker-compose.yml', and related stack files. Boundary markers: Absent. Capability inventory: Execution of 'docker build', 'docker run', 'docker-compose', and local PowerShell scripts. Sanitization: No explicit sanitization or validation of the ingested files is mentioned.
- [DATA_EXPOSURE]: Contains a hardcoded absolute file path 'C:\Users\issda\SCBE-AETHERMOORE-working' which exposes the author's local directory structure, although this is consistent with the author's provided context.
Audit Metadata