performing-ip-reputation-analysis-with-shodan

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.80). The prompt includes and encourages embedding a Shodan API key directly into code (enricher = ShodanEnricher("YOUR_SHODAN_API_KEY")), which is an insecure pattern that would require the LLM to accept and potentially emit secret API key values verbatim.

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

• Third-party content exposure detected (medium risk: 0.65). The workflow calls Shodan’s InternetDB free API at runtime (requests.get("https://internetdb.shodan.io/{ip_address}") and resp.json()), which ingests free-form JSON fields originating from an external public service (outsider-authored content) into the agent’s LLM context.