The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes a set of platform-specific tools (analysis, dashboard, analysis-meta) to query and manage data. These tools are used for their intended purpose: listing dashboards, running reports, and performing ad-hoc analysis on business metrics.
[DATA_EXPOSURE]: While the skill accesses business-sensitive LTV and revenue data, it does so within the platform's execution environment using standard queries. There are no patterns suggesting data exfiltration or the use of hardcoded credentials.
[PROMPT_INJECTION]: The skill contains instructional constraints (e.g., language policies) that are standard for agent behavior and do not attempt to bypass safety filters or override core system prompts.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it ingests data from external reports and dashboards to perform analysis. However, the risk is minimal as the data processed is structured business telemetry, and the skill lacks dangerous write or network capabilities. Severity is considered low/safe.
[DYNAMIC_EXECUTION]: The skill generates structured JSON definitions for "Revenue AI" queries based on user parameters. This dynamic generation is restricted to valid platform-specific schemas and is handled by dedicated analysis tools.

ltv-analysis