The Agent Skills Directory

[SAFE]: The skill consists entirely of instructional markdown and test cases. It does not contain any executable scripts, shell commands, or network operations.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process data from external files, which presents a surface for indirect prompt injection.
Ingestion points: Reads strategy and learning data from .agents/marketing-learnings.md, .agents/marketing-strategy.md, and various reference files in the marketing-strategy/ and ads/ directories.
Boundary markers: The instructions do not specify explicit delimiters or 'ignore' warnings for the data being read.
Capability inventory: No subprocess calls, network operations, or file-write capabilities are present in the skill files.
Sanitization: No validation or sanitization of the ingested file content is described in the instructions.
[DATA_EXPOSURE]: The skill instructions require the agent to read local sensitive business context (margins, strategy, budget constraints). This access is necessary for the skill's primary function and no mechanisms for exfiltrating this data to external domains were found.

budget-allocation