apexlang-inventory
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [SAFE]: The skill is entirely composed of documentation and metadata for application component properties and APEXlang syntax. It acts as a reference library for the AI agent to author or edit APEX application manifests.
- [NO_CODE]: There are no executable scripts (Python, Shell, JavaScript) provided within the skill; all analyzed files are Markdown or YAML configuration. While the documentation describes properties that can contain PL/SQL or JavaScript code (e.g., in
appProcessordynamicAction), these are presented as definitions for generation, not as code to be executed by the skill itself. - [INDIRECT_PROMPT_INJECTION]: As a static knowledge base, the skill does not have a runtime ingestion surface for untrusted external data that could influence the agent's internal state. It defines how an application should look, providing structure rather than acting on external inputs.
Audit Metadata