The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the GitHub CLI (gh) to list, edit, and close issues and project items on the author's repositories. It also uses ls to check for the presence of skill files on the local file system and curl to query the vendor's API.
[EXTERNAL_DOWNLOADS]: It communicates with the vendor's official domain (zynkr.ai) and GitHub to fetch and synchronize skill pipeline metadata. These operations are restricted to the author's known infrastructure.
[PROMPT_INJECTION]: As an entry-point router, the skill processes untrusted user data such as URLs, file paths, and free-form text. While it uses heuristic classification, it represents an indirect prompt injection surface when passing this data to other skills.
Ingestion points: User-supplied URLs, local file paths, and unstructured text inputs used for routing decisions.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the routing logic.
Capability inventory: Execution of shell commands via gh, curl, and ls, and the ability to auto-invoke sub-skills via the Skill tool.
Sanitization: The skill classifies input by shape but does not perform explicit sanitization or filtering before routing content to downstream skills.

zynkr