interview-synthesis

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill explicitly requires preserving and emitting verbatim participant quotes and accepts raw/pasted interview text, which means any secrets or credentials included in the input would be reproduced in the model's output (high exfiltration risk).

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

• Third-party content exposure detected (high risk: 0.70). The skill’s workflow explicitly ingests “raw notes, transcripts, @file.md references, or pasted text” and can “Preserve verbatim quotes” (i.e., arbitrary interview content provided via files/messages) into the LLM context, which could be outsider-authored free text if the operating user supplies transcripts/notes/comments/comments they didn’t author.