agentic-seo-skill

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). The presence of direct raw.githubusercontent.com links to install.sh and install.ps1 from an unknown GitHub user (Bhanunamikaze) — combined with curl | bash and irm (PowerShell) install patterns in the skill prompt — is a high-risk distribution pattern for malware, while the other listed URLs (example.com placeholders, developers.google.com, schema.org, ara.so, and the GitHub wiki) appear benign or informational.

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

• Third-party content exposure detected (high risk: 0.85). The required runtime workflow uses read_url_content / evidence scripts like fetch_page.py to fetch arbitrary URLs supplied by the operating user and ingest the fetched page text into the agent’s LLM context, which is outsider-authored free text from public web content.

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

• Potentially malicious external URL detected (high risk: 1.00). The README includes explicit commands that fetch and execute remote scripts at runtime (high-confidence examples: curl -fsSL https://raw.githubusercontent.com/Bhanunamikaze/Agentic-SEO-Skill/main/install.sh | bash and irm https://raw.githubusercontent.com/Bhanunamikaze/Agentic-SEO-Skill/main/install.ps1 followed by running install.ps1, and git clone https://github.com/Bhanunamikaze/Agentic-SEO-Skill.git plus bash install.sh), which downloads and runs external code that controls installation/runtime behavior.

MEDIUM W013: Attempt to modify system services in skill instructions.

• Attempt to modify system services in skill instructions detected (medium risk: 0.40). The skill does not ask for sudo, user creation, or edits to system/privileged configs, but it includes risky installation patterns (curl|bash installer) and explicitly recommends bypassing Windows PowerShell execution policy (Set-ExecutionPolicy ... Bypass), which weakens local security — so moderate concern.