skills/lovstudio/skills/lov-gh-access/Gen Agent Trust Hub

lov-gh-access

Pass

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes gh CLI commands to interact with GitHub APIs, interpolating user-supplied identifiers (usernames and emails) into these commands. Although it uses -f flags to pass arguments safely, the execution of privileged API calls using untrusted input is a sensitive capability.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by processing external data to drive access control decisions.
  • Ingestion points: The skill processes lists of usernames and email addresses provided by the user to identify targets for permission changes (SKILL.md).
  • Boundary markers: The instructions mandate displaying a resolution table for review and obtaining explicit user confirmation before any write operations occur.
  • Capability inventory: The skill can list, grant, and revoke repository collaborator access, including potentially assigning high-privilege roles like 'admin' or 'maintain'.
  • Sanitization: A verification step is performed via gh api "users/<login>" to ensure identifiers resolve to legitimate GitHub accounts before any modification is executed.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 24, 2026, 12:55 AM
Security Audit — agent-trust-hub — lov-gh-access