find-buyers-in-linkedin-comments
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its core functionality of ingesting and analyzing untrusted data from an external platform (LinkedIn).
- Ingestion points: The agent is instructed to read and score LinkedIn posts, comments, and user profiles to identify potential leads (
SKILL.md,references/scoring-and-selection.md). - Boundary markers: The instructions do not provide specific delimiters or ignore-instructions for the comment content, increasing the risk that the agent may accidentally execute malicious commands embedded in the text it is analyzing.
- Capability inventory: The agent has the capability to perform searches and read web content to generate reports.
- Sanitization: No explicit sanitization or filtering of the retrieved content is mentioned before the data is scored.
Audit Metadata