The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources by reading issue bodies and comments from a repository's issue tracker to manage the 'wayfinder:map'. This ingestion of potentially untrusted data creates a surface for indirect prompt injection.
Ingestion points: SKILL.md describes loading map bodies and ticket questions from an issue tracker.
Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded prompts within the ticket content.
Capability inventory: The agent has the ability to invoke subagents (/research), generate code or artifacts (/prototype), and perform domain modeling.
Sanitization: There is no evidence of sanitization or filtering applied to the issue content before processing.
[COMMAND_EXECUTION]: The skill relies on the execution of several internal vendor-provided commands and sub-skills to function, including /setup-quirk-skills, /grilling, /domain-modeling, /research, and /prototype. These are utilized for configuration and task resolution within the quantumquirkxyz ecosystem.

wayfinder