list-building
Pass
Audited by Gen Agent Trust Hub on Oct 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process data from external, untrusted sources which could contain malicious hidden instructions.
- Ingestion points: The skill (SKILL.md) instructions explicitly mention processing data from sources like 'conference attendee files', 'Google Maps scrapes', and 'LinkedIn Sales Navigator exports'.
- Boundary markers: The instructions lack specific delimiters or system-level warnings to the agent to treat external data as untrusted, increasing the risk of the agent following instructions embedded within the data.
- Capability inventory: The skill provides a local Python script (scripts/check-output.py) for deliverable validation which is intended to be executed in the agent's environment.
- Sanitization: There is no evidence of content sanitization, filtering, or validation logic to ensure that ingested text does not contain adversarial prompts.
Audit Metadata