The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill identifies a data ingestion surface by instructing the agent to read from project-specific context files. This could allow untrusted data to influence agent behavior if the context files are maliciously crafted.
Ingestion points: .agents/product-marketing-context.md and .claude/product-marketing-context.md (SKILL.md).
Boundary markers: None identified in the skill instructions.
Capability inventory: The skill documents the use of shell commands like cwebp, mogrify, jpegoptim, and curl for image optimization and verification (SKILL.md).
Sanitization: No specific sanitization or validation logic is mentioned for the ingested context data.
[COMMAND_EXECUTION]: The skill provides examples of shell commands for the agent or user to execute. These include image compression tools (cwebp, jpegoptim), batch processing (mogrify), and network requests (curl) to inspect web pages for image usage.
[EXTERNAL_DOWNLOADS]: The skill references several external services and APIs for image generation and design, including Google Gemini, Black Forest Labs (Flux), Ideogram, and OpenAI. These references target official and well-known service domains.

image