The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of the GitHub CLI (gh) to manage organizational resources. This includes creating repositories (gh repo create), managing project boards (gh project item-add), and enumerating organization-level secrets and variables (gh secret list --org firstsun-dev). These commands are necessary for the skill's primary purpose and are executed within the vendor's own organization context.
[EXTERNAL_DOWNLOADS]: The skill installs dependencies using npx skills add from the firstsun-dev/skills repository. According to security protocols, downloads from the vendor's own official repositories are considered safe. This is a standard part of the project's dependency management lifecycle.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) by ingesting untrusted data. Ingestion points: The agent reads external content from SKILLS_LIST.md and the org-wide project board README (gh project view) to derive metadata. Boundary markers: There are no explicit delimiters used during the processing of this data. Capability inventory: The skill possesses powerful capabilities including file system modification, repository creation, and secret enumeration. Sanitization: The risk is effectively mitigated by mandatory checkpoints requiring the agent to show the user exactly what it is about to do (e.g., drafted repo topics, README diffs) and wait for explicit confirmation before proceeding with irreversible actions.

firstsun-project-init