Security audit
Security checks for vulnerabilities and agentic risk
This GitHub CLI skill is coherent, but it broadly enables authenticated GitHub actions including destructive repository and issue operations without clear confirmation guidance.
Install only if you want agents to use your authenticated GitHub CLI. Before allowing it to act, require the agent to state the target owner/repository and get explicit confirmation for merges, comments, edits, transfers, deletes, archives, renames, or raw gh api calls.
The repository command list includes highly destructive actions such as delete, edit, rename, and archive with no safety warning or approval guidance. Because repository-level actions can cause irreversible loss of code, configuration changes, or service disruption across teams, exposing them plainly in a general-purpose skill makes misuse significantly more dangerous.
The manifest description says to use this skill whenever interaction with GitHub repositories, issues, pull requests, and more is needed, which is an overly broad activation condition. In an agent setting, broad routing increases the chance the skill is invoked for sensitive or destructive GitHub tasks without additional policy checks or user confirmation.
The gh issue section lists destructive capabilities such as delete and transfer alongside routine operations without any warning, approval requirement, or distinction between read-only and mutating actions. In an agent workflow, this can normalize unsafe invocation patterns and lead to accidental or unauthorized issue deletion or reassignment.
No suspicious patterns detected.
