fancy-yard-landscaper
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFENO_CODECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill does not contain any executable scripts, Python code, or Node.js logic. It consists entirely of informational and instructional markdown files.
- [COMMAND_EXECUTION]: The skill's configuration enables the
Bashtool. For a lifestyle and landscaping skill, providing access to a shell environment represents an unnecessary expansion of privileges that could be leveraged if the agent is subverted. - [PROMPT_INJECTION]: The skill has a notable indirect prompt injection surface.
- Ingestion points: Untrusted data enters the context through user-provided yard photos and property descriptions as part of the 'Photo Documentation Guide' (SKILL.md).
- Boundary markers: The instructions do not include delimiters or specific 'ignore embedded instructions' directives to prevent the agent from obeying commands hidden in analyzed content.
- Capability inventory: The agent has access to powerful tools including
Bash,WebFetch,Write, andEdit. - Sanitization: There is no evidence of input validation or sanitization instructions to handle external content safely before it is processed by high-capability tools.
Audit Metadata