find-web-developer

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly instructs the agent to discover a SERVICEGRAPH_TOKEN from the shell or .env files and to use it as a Bearer token (and to surface/save the OTP-returned token), which requires the agent to read and insert secret token values into Authorization headers or commands—an instruction that can force verbatim secret handling and thus creates exfiltration risk.

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

• Third-party content exposure detected (high risk: 0.80). The skill directly fetches and ingests firm data from the external ServiceGraph API (api.servicegraph.co) — including /v1/search and /v1/get which return firm descriptions, URLs, phones, and emails — and explicitly requires the agent to read and act on those results for shortlisting and outreach, creating a clear channel for untrusted third‑party content to influence tool use and next actions.