prospect-research-compiler
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- No Code (SAFE): The skill is composed entirely of markdown instructions. It contains no Python scripts, Node.js packages, or shell commands.
- Prompt Security (SAFE): Instructions follow standard role-play conventions for a 'prospect research expert' and do not contain jailbreak attempts or bypass instructions.
- Indirect Prompt Injection (SAFE): The skill describes ingesting data from external sources (news, social media, company websites, and financial data). While this creates a theoretical ingestion surface, the skill has no functional code or tools to perform actions, meaning it relies on the host agent's native, safe processing. 1. Ingestion points: news, social media, company websites, and financial data as specified in SKILL.md. 2. Boundary markers: Absent. 3. Capability inventory: None. 4. Sanitization: Absent.
Audit Metadata