services
Pass
Audited by Gen Agent Trust Hub on Mar 26, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill exposes the
koyeb services execcommand, which allows the agent to run arbitrary code within remote service containers. It also includes a shell wrapper (scripts/koyeb-cli.sh) that executes the CLI with passed arguments. These are appropriate administrative features for the skill's management context. - [COMMAND_EXECUTION]: Documentation in
references/koyeb-cli.mdprovides a complex shell pipeline utilizingawk,sort, andxargsto perform bulk deletion of resources. This demonstrates the skill's capability for automated resource destruction. - [EXTERNAL_DOWNLOADS]: The skill references the official
www.koyeb.comdomain for CLI installation and documentation. These are trusted vendor sources and do not represent a security risk. - [PROMPT_INJECTION]: The skill processes data from service logs and resource lists, which may contain attacker-controlled content, creating a surface for indirect prompt injection.
- Ingestion points: Data enters the agent's context through
koyeb services logsandkoyeb services listcommands. - Boundary markers: None are defined in the instructions to separate service output from agent instructions.
- Capability inventory: The skill has powerful capabilities including remote command execution (
exec) and resource deletion (delete). - Sanitization: No sanitization is performed on the output of CLI commands before it is processed by the agent.
Audit Metadata