The Agent Skills Directory

[PROMPT_INJECTION]: The skill establishes an indirect prompt injection surface by directing the agent to read and process data from project files (.agents/product-marketing-context.md or .claude/product-marketing-context.md) to guide its analysis. If these files are manipulated by an external actor, they could potentially influence the agent's output.\n- Ingestion points: SKILL.md (instructions to gather business context from marketing files).\n- Boundary markers: Absent; the instructions do not provide delimiters or warnings for the agent to distinguish between legitimate context and potentially malicious embedded instructions.\n- Capability inventory: The skill does not define restricted tools in the YAML frontmatter, thereby allowing the agent to utilize all available environment capabilities when processing the ingested content.\n- Sanitization: None; the agent is instructed to read the context files directly without specific validation or filtering steps.

pricing-strategy