The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains an indirect prompt injection attack surface because it reads formatting specifications from a remote repository's Topics/inbox/README.md and uses them to process user material.
Ingestion points: Remote README.md files from targeted repositories and user-supplied content (links, text, images) in SKILL.md workflows.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands within the remote formatting guidelines are identified.
Capability inventory: The skill utilizes gh api for remote repository writes and performs local file writes to the inbox-fallback/ directory.
Sanitization: There is no documented validation or sanitization of the content fetched from the remote repository before it influences the agent's formatting logic.

submit-to-inbox