The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection because it ingests and processes untrusted data from GitHub issue and pull request bodies.
Ingestion points: The skill uses gh issue list and gh pr list in Phase 1 to fetch the body field of issues and PRs.
Boundary markers: Absent; there are no explicit instructions to the AI to ignore instructions found within the fetched text.
Capability inventory: Bash (used for GitHub CLI commands), Write (used for saving reports to claudedocs/), and Read across the workspace.
Sanitization: Absent; the data is analyzed as-is for the purpose of categorization and cross-referencing.
[COMMAND_EXECUTION]: The skill uses the Bash tool to execute standard gh (GitHub CLI) commands for repository data collection. This is routine behavior for a triage automation skill.
[SAFE]: No obfuscation, persistence mechanisms, or credential harvesting patterns were detected. All external operations are performed via the GitHub CLI, and the output is stored locally in the workspace. The skill author 'rtk-ai' uses consistent naming conventions for the skill and its generated documentation.

rtk-triage