xy-term-crack
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions direct the AI agent to execute a local Python script (
scripts/atoms-search.py) to perform database lookups based on user-provided terms. - [EXTERNAL_DOWNLOADS]: The
scripts/atoms-search.pyscript makes network requests tohttps://api.xyskill.xyz/v1/atoms/searchto retrieve data. This domain is associated with the vendor's own infrastructure for the XY Skill system. - [INDIRECT_PROMPT_INJECTION]: The skill instructions provide a template for shell command execution that interpolates user-provided keywords directly into the command arguments (
python3 ... "<关键词>"). This creates a surface for command injection if the agent does not properly sanitize the input. - Ingestion points: User-provided business terms, sentences, or mode names provided in Phase 1 and Phase 2 of the skill execution.
- Boundary markers: No delimiters or safety warnings are provided to the agent to sanitize the keywords before shell interpolation.
- Capability inventory: Shell command execution for the
atoms-search.pyscript. - Sanitization: No sanitization or validation logic is defined in the instructions for the agent to follow.
Audit Metadata