ai-persona-creator
Pass
Audited by Gen Agent Trust Hub on Mar 5, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE]: The skill is composed strictly of Markdown documentation and templates (SKILL.md, README.md, and various files in the frameworks/ and templates/ directories) along with a JSON metadata file (prpm.json). No Python, JavaScript, or other executable code is included in the package.
- [SAFE]: A thorough analysis of all 16 files reveals no malicious patterns, such as direct prompt injections, hidden obfuscation, or persistence mechanisms. The skill facilitates a research-based workflow that uses the agent's native tools to process information according to defined psychological models.
- [PROMPT_INJECTION]: The skill defines a research phase (SKILL.md, Phase 1) where the agent is instructed to perform 32-48 web searches and save the results. This ingestion of untrusted data from external websites creates a potential surface for indirect prompt injection. The skill mitigates this risk by explicitly requiring cross-verification across at least three independent sources and several rounds of self-refinement and quality checks (SKILL.md, Step 2 and Step 8).
Audit Metadata