The Agent Skills Directory

[SAFE]: The skill is a set of text-based instructions and frameworks for a business coaching session. It does not contain executable code, script files, or remote dependencies.
[INDIRECT_PROMPT_INJECTION]: The skill has a minimal attack surface as it reads BUSINESS_CONTEXT.md to customize its advice.
Ingestion points: Reads BUSINESS_CONTEXT.md from the project root or current directory.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the skill's logic.
Capability inventory: The skill only performs file read operations and file write operations (saving to the reviews/ directory). It lacks network access or arbitrary command execution capabilities.
Sanitization: No explicit sanitization or validation of the context file's content is defined, but the lack of dangerous tools mitigates the risk.
[DATA_EXPOSURE]: The skill reads business-related metadata (team size, revenue, roles) to provide context. This data is used only to generate a local report and is not exfiltrated to any external service.

delegation