hexstrike-forge

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). It explicitly asks the agent to request user credentials for authenticated tests and to include "any credentials or context the next session needs" in handoff, which requires the LLM to receive and potentially output secrets verbatim, posing exfiltration risk.

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

• Third-party content exposure detected (low risk: 0.10). Outsider-authored free text is only ingested via the user-provided target list/scan prompt, while the skill’s runtime reads tool outputs from the target environment (not a public feed/queue where arbitrary third parties can submit text that the workflow monitors without selecting a specific item).