assess-me

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly collects local "prompts" and instructs the agent to anchor claims to specific evidence including "a verbatim quote" from transcripts/current session, which forces copying any secrets present in those transcripts into the generated assessment (and thus output) — an exfiltration risk.