contract-drafting
Pass
Audited by Gen Agent Trust Hub on Oct 31, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONMETADATA_POISONING
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided project details and technical specifications to generate Statements of Work (SOWs) and other agreements. While this creates a surface for indirect prompt injection, the skill is strictly limited to text generation and lacks high-privilege capabilities such as file system writes, network access, or shell command execution.
- [METADATA_POISONING]: The
SKILL.mdfrontmatter contains alast_revieweddate set in the future (2026-09-06). While this metadata is technically inaccurate, it does not appear to be used for deceptive purposes or to bypass security logic.
Audit Metadata