The Agent Skills Directory

Data Ingestion Surface: The skill is designed to process data from external connectors and user-uploaded files (CSV, XLSX). This represents a surface for indirect prompt injection where instructions embedded in data could potentially influence agent behavior.
Ingestion points: Data is ingested via uploaded CSV/XLSX files and connected data sources as described in SKILL.md.
Boundary markers: The instructions do not specify explicit boundary markers or delimiters to isolate data from instructions.
Capability inventory: The skill has the capability to fetch web content (WebFetch), read files (Read), and invoke other skills (report-builder, business-pulse, cash-flow-snapshot, invoice-chase).
Sanitization: There are no explicit instructions for sanitizing or validating the content of the ingested files.
Orchestration of Sub-Skills: The skill orchestrates multiple sub-skills to aggregate and summarize business data. This involves passing data between tools to create a unified report, which is a standard pattern for complex agent tasks.
External Resource Referencing: The skill references a shared resource (../../shared/artifact-style.md) to apply visual styling to its HTML artifacts. Utilizing shared templates is a common practice for maintaining consistency across an agent's output.
Automated Delivery and Scheduling: The skill implements a scheduling mechanism to deliver reports periodically. This functionality is presented as a user-facing feature to automate recurring tasks.

report-pack