product-differentiation-tiktok
Pass
Audited by Gen Agent Trust Hub on Oct 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external text data (product reviews and comments) which could theoretically contain adversarial instructions designed to influence an LLM.
- Ingestion points: The
competitor_reviewsandmy_reviewsdata processed by theextract_pain_pointsandextract_selling_pointsfunctions inscripts/analyzer.py. - Boundary markers: Not present in the script's internal logic.
- Capability inventory: None. The script relies on standard Python libraries and does not have the capability to write files, perform network operations, or execute shell commands based on the processed data.
- Sanitization: The script utilizes literal keyword matching against static dictionaries (
PAIN_POINT_KEYWORDS,SELLING_POINT_KEYWORDS), which effectively filters out any non-matching content including potential injection commands. - [EXTERNAL_DOWNLOADS]: The
SKILL.mdfile provides installation instructions usingnpxto download a package from the author's official repository. This is a standard deployment method for this vendor's tools and points to their own infrastructure. - [SAFE]: No obfuscation, command execution, or data exfiltration patterns were identified. The logic in
scripts/analyzer.pyis transparent and restricted to statistical analysis and report generation.
Audit Metadata