persona-builder
Pass
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to ingest and analyze untrusted external data from LinkedIn profiles, user posts, and job descriptions to build detailed personas. If an attacker controls a profile or post the agent researches, they could embed malicious instructions.
- Ingestion points: External content from LinkedIn profiles, activity streams, and job boards are ingested into the agent context via MCP as described in the 'Process' section of
SKILL.md. - Boundary markers: The instructions do not define specific delimiters or include warnings to ignore potentially embedded instructions within the retrieved external content.
- Capability inventory: The skill is designed for data collection and marketing persona generation. It does not exhibit capabilities for sensitive file access, command execution, or persistent network exfiltration.
- Sanitization: There is no documented requirement for the agent to sanitize or validate the external content before processing it for persona creation.
Audit Metadata