skills/patsnap/skills/mab-fto-check/Gen Agent Trust Hub

mab-fto-check

Pass

Audited by Gen Agent Trust Hub on Aug 27, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a bundled Python script (scripts/mab_fto_recall_estimator.py) to perform statistical calculations (Chapman capture-recapture estimation) for search recall. This is a legitimate use of local command execution for a provided utility script.
  • [EXTERNAL_DOWNLOADS]: The skill retrieves patent information from eureka.zhihuiya.com and directs users to open.zhihuiya.com for authentication. These are official domains associated with the skill's author (Patsnap) and are required for the tool's core functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted user input in the form of biological sequences (e.g., VH, VL, CDR regions). Although this represents a potential attack surface where instructions could be hidden in sequence data, the skill uses these inputs primarily as parameters for specialized search tools (ls_sequence_alignment), which significantly limits the risk of the agent executing arbitrary commands from the data.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 27, 2026, 01:22 AM