eve-repo-upkeep
Warn
Audited by Gen Agent Trust Hub on Feb 19, 2026
Risk Level: MEDIUMCOMMAND_EXECUTION
Full Analysis
- Command Execution (MEDIUM): The skill directs the agent to execute shell commands for container builds (
docker compose up --build) and environment deployments (eve env deploy). These commands operate on the state of the repository which the agent has just modified, potentially executing malicious logic if the repository configuration was subverted. - Indirect Prompt Injection (LOW): This skill has a significant attack surface for indirect injection. Ingestion points: Processes multiple local files including
.eve/manifest.yaml,skills.txt,AGENTS.md, andagents/configuration files. Boundary markers: The skill lacks explicit instructions to ignore or delimit embedded instructions within these data files. Capability inventory: The agent can execute system commands (docker,eveCLI) and write to the local filesystem. Sanitization: There is no evidence of sanitization or structural validation for the content of the files before they are used to influence the deployment process.
Audit Metadata