sasjs-framework
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
- [NO_CODE]: The skill consists purely of Markdown documentation and a JSON schema file. It does not contain any executable scripts, shell commands, or automation logic.
- [INDIRECT_PROMPT_INJECTION]: The skill documentation explicitly addresses the risk of SAS macro injection from untrusted web inputs entering the agent context. It provides remediation guidance for these surfaces:
- Ingestion points: Web request values (input tables, URL params) described in SKILL.md.
- Boundary markers: Instructions include explicit abort mechanisms (%mp_abort) and delimiter handling.
- Capability inventory: Mentions SAS platform capabilities such as file creation (%mv_createfile) and macro execution.
- Sanitization: Provides code patterns for input validation using %mp_validatecol and whitelist-based lookup to prevent command injection into generated SAS code.
- [SAFE]: All referenced URLs, repositories, and documentation links belong to the official vendor domains (sasjs.io, sasjs.com) or are well-known developer services.
Audit Metadata