signal-sourcer

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill shows a CLI pattern that embeds an API key inline (COLDIQ_API_KEY= npx -y @coldiq/mcp@latest), which encourages including secrets verbatim in commands and could require the model to handle or reproduce secret values.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 0.90). This skillpack contains deliberate, privacy-invasive instructions (scraping review sites and LinkedIn, installing person-level tracking pixels, webhooks that push PII to third parties) and guidance to hide that tracking from prospects — patterns that enable targeted data exfiltration and deceptive outreach.

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

• Potentially malicious external URL detected (high risk: 0.80). The skill repeatedly instructs connecting to ColdIQ (https://coldiq.com/marketplace and https://api.coldiq.com) and includes an explicit setup command that fetches and runs remote code at runtime via COLDIQ_API_KEY=<key> npx -y @coldiq/mcp@latest, so the skill relies on executing externally fetched code (ColdIQ MCP) during runtime.