pr-threads-resolve

Warn

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill requires the installation of a third-party GitHub CLI extension agynio/gh-pr-review. This involves downloading and installing executable code from an external repository that is not part of the trusted vendors list.
  • [COMMAND_EXECUTION]: The skill utilizes potentially dangerous shell piping patterns, such as jq -r '.threads[].id' | xargs -I {} gh pr-review threads resolve --thread-id {}. This pattern executes multiple subcommands based on data retrieved from an external source (GitHub PR threads).
  • [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection by processing Pull Request review comments which are user-controlled external data.
  • Ingestion points: Uses gh pr-review threads list and gh pr-review review view to fetch thread content and comments from GitHub (SKILL.md).
  • Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present in the provided CLI examples.
  • Capability inventory: The skill can execute shell commands, pipe output to xargs, and perform write operations to GitHub (resolving threads).
  • Sanitization: Relies on jq for parsing structured JSON, which mitigates some risks, but the resulting IDs are passed directly to shell commands via xargs.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 5, 2026, 03:16 AM