claude-code-marketing-skills

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.70). Most links are standard (GitHub repo, company pages, docs) and generally low-risk, but the explicit direct shell installer (https://raw.githubusercontent.com/.../install.sh) and the provided curl | bash install pattern is a high‑risk distribution vector because it downloads and executes remote code from an external repo, so treat the bundle as moderately high risk.

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

• Third-party content exposure detected (high risk: 1.00). The skill explicitly uses WebFetch/WebSearch to ingest public website content (see "Free skills" and examples like /seo-audit https://example.com, /landing-page-review https://example.com/landing, and /competitor-analysis) so the agent reads and acts on untrusted third‑party web content as part of its workflow, which could be used for indirect prompt injection.

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

• Potentially malicious external URL detected (high risk: 1.00). The quick-install command fetches and pipes remote code to a shell—curl -sSL https://raw.githubusercontent.com/cognyai/claude-code-marketing-skills/main/install.sh | bash—which downloads and executes code (and the repo https://github.com/cognyai/claude-code-marketing-skills provides the skill files the agent loads), so this is a runtime external dependency that executes remote code and controls agent behavior.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (high risk: 1.00). The skill includes explicit integrations with advertising APIs (Google Ads, Meta Ads, LinkedIn Ads) via Cogny MCP and a "Cogny Agent (Full Autonomy)" that "executes approved changes (with confirmation)". Managing/updating ad campaigns, bids, and budgets is supported by the listed premium capabilities (e.g., LinkedIn Ads Audit, LinkedIn Micro Campaigns, and the architecture listing Google/Meta/LinkedIn Ads APIs). Because it can connect to ad platforms and perform changes (which can include updating spend/budgets) via their APIs, this constitutes direct financial execution capability for ad spend management.