The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes uv run lintro and various gh api calls to manage repository states, rulesets, and issues.
[COMMAND_EXECUTION]: Dynamically generates and runs Python scripts that interface with the gh CLI to create milestones, epics, and issues from a structured specification.
[EXTERNAL_DOWNLOADS]: Fetches external CI/CD workflows (Winnow) and standard community configuration files (e.g., .editorconfig, SECURITY.md, Renovate, and gitleaks configs) to populate the target repository.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface by ingesting and processing untrusted data from external sources.
Ingestion points: GitHub API responses (repository rulesets) and content of fetched workflow/configuration files.
Boundary markers: None detected; the agent does not use specific delimiters to isolate ingested external data from its internal instructions.
Capability inventory: Subprocess execution (uv, gh, python), file system writes, and repository modification capabilities.
Sanitization: No explicit sanitization or validation of external content or API metadata is performed before use in script generation or backlog creation.

properize