linkup
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [External Downloads] (SAFE): Installation is performed via the
reflyCLI, which is the expected method for this platform. - [Data Exposure] (SAFE): No sensitive information, such as API keys or system credentials, is hardcoded or exposed.
- [Indirect Prompt Injection] (SAFE): The skill ingests LinkedIn profile data, creating a surface for potential indirect prompt injection. However, this is inherent to its search function, and its capabilities are restricted to displaying retrieved results.
- Ingestion points: LinkedIn profiles via
refly workflow toolcalls(SKILL.md). - Boundary markers: None identified.
- Capability inventory:
reflyCLI for workflow management and data extraction. - Sanitization: Not explicitly mentioned in the skill definition.
Audit Metadata