The Agent Skills Directory

[SAFE]: The skill consists entirely of markdown instructions, templates, and workflow descriptions for financial analysis. There are no executable scripts, shell commands, or obfuscated payloads.
[NO_CODE]: No source code, scripts, or external dependencies (Python or Node.js) are included or referenced for installation.
[PROMPT_INJECTION]: The skill includes instructions to process data from external ERP and BI systems. While this presents a surface for indirect prompt injection if those systems contain malicious data, it is a standard functional requirement for the skill's stated purpose.
Ingestion points: ERP/BI system data (SKILL.md).
Boundary markers: None specified; the skill assumes direct processing of retrieved data.
Capability inventory: Report generation and structured data analysis.
Sanitization: No explicit sanitization or input validation logic is provided for the data being ingested.
[DATA_EXPOSURE]: The skill references a relative path ../../CLAUDE.md to load project-level configuration and formatting standards. This is a common pattern in development-oriented agent environments and does not constitute a security risk in this context.

financial-report-generation