personality-profiler
Pass
Audited by Gen Agent Trust Hub on Feb 21, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- PROMPT_INJECTION (LOW): The skill is vulnerable to indirect prompt injection through the analysis of raw text from social media data exports. Ingestion points: Data enters the agent via scripts/parse_twitter.py (tweets.js), scripts/parse_linkedin.py (Shares.csv), and scripts/parse_instagram.py (posts_1.json). Boundary markers: Absent; the skill lacks explicit instructions to treat the analyzed content as untrusted or to use delimiters to prevent command execution. Capability inventory: The agent performs local file reading and writing using standard Python libraries; scripts contain no network or shell execution capabilities. Sanitization: Absent; the scripts extract and normalize text for character encoding but do not filter for malicious prompt instructions.
Audit Metadata