sysadmin
Warn
Audited by Gen Agent Trust Hub on Aug 30, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONPRIVILEGE_ESCALATIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill defines an 'Operator' role responsible for running commands directly on a machine to manage its configuration, as documented in the git repository.
- [PRIVILEGE_ESCALATION]: The instructions specifically guide the agent on the use of sudo for administrative tasks, providing guidance on how to correctly expand paths when switching users.
- [INDIRECT_PROMPT_INJECTION]: The agent is instructed to automatically read and apply instructions found in external files, which could contain malicious directives.
- Ingestion points: The skill processes instructions from markdown files located in the docs/handoff/ directory (SKILL.md).
- Boundary markers: There are no explicit instructions or delimiters defined to separate the handoff file content from the system prompt or to warn the agent about potentially malicious instructions embedded in these files.
- Capability inventory: The skill permits full command execution on the host machine, including file writing to .reports/ and docs/, and privilege escalation via sudo for administrative changes.
- Sanitization: There is no requirement for the agent to sanitize, validate, or filter the instructions found in the handoff files before execution.
Audit Metadata