bypass-crossplag
Pass
Audited by Gen Agent Trust Hub on Oct 25, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process user-provided drafts for rewriting, which introduces an indirect prompt injection surface. Malicious instructions embedded within the user's text could theoretically influence the agent's behavior during the rewriting process.\n
- Ingestion points: User-provided text drafts for humanization (SKILL.md).\n
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are provided.\n
- Capability inventory: The skill references the use of external tools like
@rephrasy/mcpfor processing.\n - Sanitization: No mention of input sanitization or validation logic.\n- [EXTERNAL_DOWNLOADS]: The instructions refer to the
@rephrasy/mcpNode.js package available on the public NPM registry.\n - Package:
@rephrasy/mcpavailable viahttps://www.npmjs.com/package/@rephrasy/mcp.
Audit Metadata