sankhya-dashboard-html-jsp-custom-best-pratices

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze and generate JSP and SQL code provided by the user. This creates an ingestion point for potentially malicious instructions embedded in comments or metadata within the user's project files. * Ingestion points: The skill processes user-provided JSP, HTML, and SQL files as part of its core purpose. * Boundary markers: There are no explicit delimiters or warnings to the agent to ignore instructions embedded in the analyzed code. * Capability inventory: The skill utilizes code generation and UI modification capabilities but does not provide access to network exfiltration or direct file system writes. * Sanitization: While the skill promotes sanitizing SQL parameters at runtime, it does not specify sanitization of the input source code against prompt injection.
  • [SAFE]: The skill provides strong defensive coding recommendations, including the use of JSTL tags instead of Java scriptlets, parameter sanitization for SQL queries using fn:replace, and the avoidance of hardcoded credentials or sensitive URLs.
  • [SAFE]: The installation instructions in the README are transparent and describe standard local configuration procedures for AI agent skills using symlinks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 01:59 PM
Security Audit — agent-trust-hub — sankhya-dashboard-html-jsp-custom-best-pratices