voice-matching-wizard
Pass
Audited by Gen Agent Trust Hub on May 20, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill is composed of purely instructional Markdown files (
SKILL.md,WIZARD.md,README.md) and templates. It does not perform any network operations, file system modifications, or command executions. - [INDIRECT_PROMPT_INJECTION]: The skill's primary function involves analyzing untrusted data in the form of user-provided writing samples. While this is an ingestion surface for indirect prompt injection, the skill's activities are limited to pattern extraction and text generation without any risky capabilities like code execution or network exfiltration.
- Ingestion points: User-provided writing samples requested in
SKILL.md(Phase 1) andWIZARD.md(Path A/B). - Boundary markers: None; the instructions suggest pasting samples directly.
- Capability inventory: None; the skill is restricted to analyzing text and producing a new Markdown file.
- Sanitization: None; the skill relies on the underlying LLM's inherent safety guardrails when processing the content.
Audit Metadata