The Agent Skills Directory

[COMMAND_EXECUTION]: The skill relies on the GitHub CLI (gh) to perform repository operations. It uses commands like gh repo list to discover forks and gh repo sync to update them. These commands are executed as subprocesses and are central to the skill's stated purpose.
[CREDENTIALS_UNSAFE]: The skill manages GitHub authentication. It supports using a Personal Access Token (PAT) via the command echo "<TOKEN>" | gh auth login --with-token. While this handles sensitive credentials, it uses the official CLI's recommended method for non-interactive login.
[EXTERNAL_DOWNLOADS]: The skill includes references to official GitHub documentation and the GitHub CLI installation page (https://cli.github.com/). These are well-known, trusted sources used for setup and reference purposes.
[PROMPT_INJECTION]: The instructions include clear logic for handling user requests and environment validation without attempting to override the agent's underlying safety protocols or system instructions.
[INDIRECT_PROMPT_INJECTION]: The skill has a potential attack surface by processing data from an external source.
Ingestion points: Data is ingested from the GitHub API via gh repo list and gh api commands as described in SKILL.md.
Boundary markers: There are no explicit delimiters or boundary markers used when processing the list of repository names.
Capability inventory: The agent has the capability to execute shell commands (gh repo sync, gh api) and handle authentication (gh auth login) as seen in SKILL.md.
Sanitization: No explicit sanitization or validation of repository names is performed before they are passed into shell commands, though the risk is naturally mitigated by GitHub's strict repository naming conventions.

github-fork-sync-assistant