ux-copy
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- Instructional Design: The skill provides structural guidance for writing UX copy. It functions as a set of logical prompts and does not include scripts, binaries, or command-line operations.
- Data Ingestion Surface: The skill allows the agent to ingest content from user input and connected external tools like design platforms and knowledge bases. While this introduces a surface for indirect prompt injection where instructions could be hidden in external data, the skill is evaluated as safe because it lacks the necessary capabilities—such as network access or file system write permissions—to execute malicious actions based on those instructions.
- Vulnerability Surface Analysis (Indirect Prompt Injection):
- Ingestion points: User input via the $ARGUMENTS variable and external data fetched through connectors in SKILL.md.
- Boundary markers: The skill does not utilize specific delimiters or instructions to isolate untrusted data.
- Capability inventory: No subprocess calls, network requests, or file-writing operations are present in the skill files.
- Sanitization: No explicit sanitization or filtering logic is defined.
Audit Metadata