docs-skill-builder

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). The GitHub URL points to a user repository (which may be legitimate but must be vetted for reputation/activity), while the wildcard patterns (https://* and http://*) are unconstrained and could resolve to any site — including typosquatters, direct executable hosts, or malicious redirects — so taken together they represent an untrusted/suspicious download source.

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

• Third-party content exposure detected (high risk: 0.95). The required runtime workflow fetches and mirrors outsider-authored free text from either public web pages (curl in fetch_web_source when SOURCE_URL is web+https://...) or from an external Git repository (git clone then cp of *.md/*.txt/... into skills/<skill>/references/), which becomes readable prose in the agent’s LLM context via the installed skill’s references/ files.