The Agent Skills Directory

[SAFE]: The skill utilizes the official GitHub CLI (gh) for all authentication and API interactions, ensuring that credentials are managed securely by the underlying operating system and the official tool rather than being hardcoded or stored by the skill itself.
[COMMAND_EXECUTION]: The skill provides clear templates for shell commands using gh, jq, and yq. It explicitly implements a safe GraphQL execution pattern in patterns/graphql-execution.md that uses temporary files to prevent shell variable expansion vulnerabilities.
[PROMPT_INJECTION]: The skill processes external data from GitHub, such as issue titles and PR descriptions, which constitutes an indirect prompt injection surface. However, the risk is minimal as the instructions focus on using jq to filter structured data for reporting rather than re-processing text as new instructions.
Ingestion points: Data retrieved via gh api calls in references/cross-repo-search.md and references/team-analytics.md.
Boundary markers: None explicitly defined for untrusted data interpolation.
Capability: Execution of gh CLI commands with organization-level scopes.
Sanitization: Use of jq to select specific JSON fields, which naturally filters out non-selected metadata.

github-multi-repo-manager