cv-generator
Pass
Audited by Gen Agent Trust Hub on Oct 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from LinkedIn profiles, personal portfolio websites, and GitHub profiles. This creates a vulnerability surface where a malicious actor could embed instructions within these sources to manipulate the agent's behavior during CV generation.
- Ingestion points: Source Selection section (LinkedIn URL, Portfolio URL, LinkedIn PDF, and GitHub profile).
- Boundary markers: The instructions do not define any delimiters or specific warnings to ignore commands embedded within the ingested data.
- Capability inventory: Web fetching and document parsing are utilized to extract content from external sources.
- Sanitization: No sanitization, filtering, or escaping logic is described for the content retrieved from these external inputs.
- [EXTERNAL_DOWNLOADS]: The skill facilitates the automated fetching of content from user-provided external URLs, specifically targeting LinkedIn profiles and personal portfolio websites for data extraction.
Audit Metadata