fresh-graduate-resume-optimizer
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes user-supplied resume descriptions, creating an untrusted data ingestion surface. While a user could include instructions within the resume text to deviate the agent from its persona, the impact is strictly limited to text output quality. The skill lacks dangerous capabilities such as network access or file system modifications.
- Ingestion points: User-supplied campus, internship, and project descriptions processed by SKILL.md.
- Boundary markers: Workflow steps define the task, but the prompt lacks explicit delimiters (like XML or markers) to isolate user input from instructions.
- Capability inventory: The skill is restricted to text generation and does not utilize any shell, network, or file-writing tools.
- Sanitization: No input validation or instruction filtering is present.
Audit Metadata