sysadmin
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill defines a protocol where the agent (as an 'Operator') reads and applies tasks from
docs/handoff/files. This creates an indirect prompt injection surface where instructions placed in those files could influence the agent's actions on the machine. - Ingestion points: The agent is instructed to process task files located in
docs/handoff/(SKILL.md). - Boundary markers: Absent; there are no instructions to the agent to treat the content of handoff files as data rather than instructions, nor are there specific delimiters defined.
- Capability inventory: High; the 'Operator' role is explicitly empowered to run commands that change the machine configuration and commit changes to git.
- Sanitization: Absent; the skill relies on the agent's manual review ('Think first') and user confirmation for unexpected events rather than technical sanitization or validation of handoff file content.
Audit Metadata