rightcode-setup-skills

Pass

Audited by Gen Agent Trust Hub on May 15, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill is a prompt-driven utility that scaffolds configuration files (docs/agents/*.md) and updates project manifests (CLAUDE.md or AGENTS.md).
  • [COMMAND_EXECUTION]: Instructions guide the agent to run git remote -v and read .git/config to identify the project's hosting provider (GitHub, GitLab, or local). This is a standard discovery process for a setup tool.
  • [SAFE]: The skill enforces a human-in-the-loop workflow, explicitly stating that it must present findings, walk the user through decisions, and show a draft for editing before performing any write operations.
  • [SAFE]: Uses disable-model-invocation: true in the frontmatter, ensuring that the model cannot autonomously execute tools and must rely on a prompt-driven interaction with the user.
  • [SAFE]: External references (e.g., links to the GitLab CLI) target well-known, official services and do not involve unverified downloads or remote code execution.
Audit Metadata
Risk Level
SAFE
Analyzed
May 15, 2026, 10:55 AM
Security Audit — agent-trust-hub — rightcode-setup-skills