The Agent Skills Directory

[SAFE]: The skill is a collection of educational Markdown files that provide a theoretical framework for marketing offer design. All content is informational and intended to guide an AI agent's reasoning processes for marketing tasks.
[NO_CODE]: There are no scripts (Python, Node.js, Shell), binaries, or command-line instructions within any of the skill files. The skill relies entirely on natural language instructions and structured references.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest local context from files such as .agents/marketing-strategy.md to personalize its marketing advice.
Ingestion points: instructions in SKILL.md prompt the agent to read marketing strategy and learnings files from .agents/, .claude/, or the root directory.
Boundary markers: The skill does not specify explicit delimiters or "ignore instructions" warnings for these ingested files.
Capability inventory: None. A review of all 9 files confirms there are no tool calls, network operations, file-write capabilities, or subprocess execution patterns available to be exploited.
Sanitization: Not applicable as the ingested data is used only as informational context for text generation and does not influence executable actions.

offers