The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the gh CLI and Linear MCP tools to perform standard project management operations.
It executes commands such as gh issue create, gh issue edit, and gh api to manage GitHub issues and sub-issues.
It interacts with the Linear API through MCP tools (e.g., save_issue, list_issues) to track work items.
These commands are restricted to the context of the project's issue tracker and align with the skill's documented purpose.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection surface due to its interaction with untrusted external data from issue trackers.
Ingestion points: The skill reads issue titles, descriptions, and comments from external trackers (GitHub and Linear) in references/trackers.md.
Boundary markers: The skill does not implement explicit delimiters or boundary markers to separate external content from agent instructions.
Capability inventory: The skill can execute shell commands (via gh), write to the local filesystem (.scratch/ directory), and perform network operations through the issue tracker APIs.
Sanitization: No explicit sanitization or filtering of external issue content is documented before the agent processes the data.

wayfinder-plus