The Agent Skills Directory

[SAFE]: The skill is a project management aid designed to interact with the GitHub CLI (gh) for milestone tracking. Its operations are transparent and consistent with its described functionality.
[COMMAND_EXECUTION]: The skill instructs the agent to execute standard GitHub CLI commands, specifically gh milestone list and gh milestone create. These commands are used to inspect existing project states and record planning outcomes as authorized by the user during the interactive session.
[INDIRECT_PROMPT_INJECTION]: The skill ingests data from local project files, such as README.md, package.json, and design documents, to provide context for the planning process. While this represents a potential surface for indirect prompt injection, the risk is minimal as the agent is instructed to summarize its findings and wait for explicit operator confirmation before defining or creating milestones.
Ingestion points: Project root files (README.md, CLAUDE.md, package.json) and prototype folders (design/, mockups/, etc.).
Boundary markers: Absent; however, the workflow requires manual operator approval of the summarized context.
Capability inventory: Read access to local project files and write access to GitHub milestones via the gh CLI.
Sanitization: None; the agent relies on interactive human feedback to validate interpreted data.

plan-epic