direction-attribute
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides legitimate educational content focused on web accessibility and internationalization (i18n). The provided code snippets and CSS/JavaScript examples represent best practices for RTL support and do not contain hidden malicious logic or suspicious external dependencies.
- [INDIRECT_PROMPT_INJECTION]: The skill establishes an attack surface by instructing the agent to ingest and analyze external code (templates, rendered HTML, and components). This is a standard requirement for the skill's primary purpose of code review.
- Ingestion points: In
SKILL.md, the "Check" and "Code Review" sections instruct the agent to verify HTML documents, templates, and server-rendered markup. - Boundary markers: The skill does not specify markers to delimit untrusted project content or provide warnings to the agent about ignoring instructions embedded within the analyzed code.
- Capability inventory: The skill does not request or implement any high-risk capabilities such as network exfiltration, sensitive file system access (beyond reading the files to be reviewed), or subprocess execution.
- Sanitization: The instructions do not specify any sanitization or filtering of the content being reviewed.
Audit Metadata