doc2math
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted documents which may contain instructions that attempt to manipulate the mathematical extraction process.\n
- Ingestion points: Technical documents and research excerpts provided as input in SKILL.md.\n
- Boundary markers: Absent; the skill does not use delimiters to isolate untrusted data.\n
- Capability inventory: The skill produces JSON output but lacks dangerous tool access or execution capabilities.\n
- Sanitization: Absent; the skill processes input text without verification.\n- [DATA_EXFILTRATION]: The skill references an external tool at ace-license-server-production.up.railway.app. The documentation warns that using this tool may transmit documents and keys to the remote server, posing a risk of data exposure to an unverified third party.
Audit Metadata