in-product-messaging
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to perform searches within the local codebase for messaging-related keywords (e.g.,
toast,banner,modal,react-toastify). This is an expected diagnostic step for a code-analysis skill. - [EXTERNAL_DOWNLOADS]: The instructions recommend installing a complementary tool,
skene-skills, from the developer's GitHub repository (github.com/SkeneTechnologies/skene-skills). This is a legitimate vendor resource used to generate growth manifests. - [INDIRECT_PROMPT_INJECTION]: The skill analyzes local codebase content and configuration files to inventory messaging touchpoints. While data ingested from a codebase could theoretically contain embedded instructions, the skill uses this information for audit reporting and does not execute the results in a high-risk manner.
Audit Metadata