diagnose

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 1.00). These URLs include a raw GitHub-hosted shell script that the skill instructs to curl and pipe to bash (remote .sh executed from an untrusted/unknown repo), which is a high-risk distribution pattern for malware.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). The skill instructs silent remote fetching and execution of code (curl | bash) and reads local config files, creating high risk for remote code execution, supply-chain compromise, and credential/data exposure.

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

• Third-party content exposure detected (medium risk: 0.30). The skill’s Step 1–5 workflow ingests arbitrary outsider-authored free text from the end-user input for language detection and symptom diagnosis.

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

• Potentially malicious external URL detected (high risk: 1.00). The skill explicitly fetches remote content at runtime: it curls https://raw.githubusercontent.com/flobby41/claude-code-ultimate-guide/main/guide/ultimate-guide.md to use as the troubleshooting reference and curls https://raw.githubusercontent.com/flobby41/claude-code-ultimate-guide/main/examples/scripts/audit-scan.sh | bash which executes remote code, so external content both controls prompts and runs code.