codeguard

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). Suspicious — the links point to an unknown/personal GitHub repo and raw install scripts (install.sh / install.ps1) which are direct executable scripts you might be tempted to curl|bash or iwr|iex; running them would execute arbitrary code from an unverified author (and the yourdomain.com entry is just a placeholder/unknown domain), so treat as high risk until the repository and author are verified.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow ingests outsider-authored free text from the operating user’s project files (e.g., source code, tests, .env contents) by reading them as text during the detection layer (python scripts/codeguard_workflow.py scan --path "<project>" → scan_project() / run_security_scan() / run_cost_analysis() / run_test_analysis() read arbitrary file contents and then embed findings into generated Markdown that is fed back into the agent context).

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 1.00). The README and install wrappers instruct fetching and executing remote installer code (e.g., curl -fsSL https://raw.githubusercontent.com/yousefabdallah171/code-quality-reviewer/main/install.sh | bash, iwr -useb https://raw.githubusercontent.com/yousefabdallah171/code-quality-reviewer/main/install.ps1 | iex, and git clone https://github.com/yousefabdallah171/code-quality-reviewer.git), which downloads and runs remote code during installation, so this is a high-confidence runtime-executed external dependency.