The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by instructing the agent to ingest content from project-specific marketing context files.
Ingestion points: Instructions in SKILL.md direct the agent to read .agents/product-marketing.md, .claude/product-marketing.md, or product-marketing-context.md at the start of the interaction.
Boundary markers: There are no boundary markers or instructions to treat the content of these files as data rather than instructions.
Capability inventory: The skill utilizes shell commands for image optimization (cwebp, mogrify, jpegoptim) and web scraping (curl).
Sanitization: The skill lacks mechanisms to sanitize or validate the content ingested from the marketing context files before it influences agent operations.
[COMMAND_EXECUTION]: The skill provides example shell commands for image processing and network verification.
Evidence includes commands like cwebp -q 80 input.png -o output.webp, mogrify -format webp -quality 80 *.png, jpegoptim --max=80 --strip-all *.jpg, and a curl pipeline to inspect remote image sources.

image