portainer-skill
Warn
Audited by Gen Agent Trust Hub on Feb 19, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- COMMAND_EXECUTION (MEDIUM): The skill directs the agent to execute shell-based
curlcommands. The use of variables like{container_id}and{action}within these commands poses a risk of command injection if those values are sourced from untrusted or malicious container metadata.\n- DATA_EXFILTRATION (LOW): Accessing container logs (/logs) introduces a data exposure risk, as logs frequently contain sensitive runtime information such as secrets or environment variables which are then ingested into the AI's context.\n- PROMPT_INJECTION (LOW): The skill is susceptible to indirect prompt injection because it processes untrusted data from container logs.\n - Ingestion points: Container logs are ingested through the
curltemplates defined inSKILL.md.\n - Boundary markers: Absent. No delimiters or instructions are provided to the agent to distinguish log content from system instructions.\n
- Capability inventory: The skill possesses the ability to execute shell commands (
curl) and modify container states.\n - Sanitization: Absent. The skill lacks mechanisms to sanitize or validate the content of the logs before they are interpreted by the agent.
Audit Metadata