li-inbox
Pass
Audited by Gen Agent Trust Hub on Sep 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from LinkedIn messages and screenshots provided by the user. An attacker could embed malicious instructions within a LinkedIn message to influence the agent's triage logic or the content of generated drafts.
- Ingestion points: User-provided LinkedIn message text and screenshots as described in the Input section of SKILL.md.
- Boundary markers: None specified. The skill does not use delimiters or instructions to ignore embedded commands within the message content.
- Capability inventory: The skill drafts text replies based on sorted buckets but does not have capabilities for network operations or file system modifications.
- Sanitization: There is no evidence of filtering, escaping, or validation of the untrusted input data.
Audit Metadata