stakeholder-summary
Pass
Audited by Gen Agent Trust Hub on Sep 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external contract reviews and legal memos as its primary input, creating a surface for indirect prompt injection where malicious instructions embedded in the source documents could attempt to influence the agent's output.
- Ingestion points: Reads contract reviews, legal findings, and context from workspace files like CLAUDE.md.
- Boundary markers: No specific delimiters or explicit instructions to disregard embedded commands in the source data are provided.
- Capability inventory: The skill is limited to text summarization and formatting; it does not contain capabilities for network operations, file system modifications, or command execution.
- Sanitization: No input validation, escaping, or filtering of the legal source text is implemented prior to summarization.
Audit Metadata