xy-close
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a provided Python script (
scripts/atoms-search.py) to search a database of sales strategies and case studies. This is a primary functional feature of the skill. - [EXTERNAL_DOWNLOADS]: The search script performs network requests to a vendor-controlled API (
https://api.xyskill.xyz) to retrieve search results. This is a documented retrieval mechanism and does not interact with sensitive user data. - [INDIRECT_PROMPT_INJECTION]: The skill retrieves data from external files or a remote API and integrates it into the conversation context. This presents a potential surface for indirect prompt injection.
- Ingestion points: Data returned by the
scripts/atoms-search.pytool. - Boundary markers: The instructions in
SKILL.mdguide the agent to cite source IDs but do not employ strict delimiters for the retrieved content. - Capability inventory: The agent can execute the
scripts/atoms-search.pyscript and read local knowledge files via that script. - Sanitization: No specific sanitization or filtering of the content retrieved by
scripts/atoms-search.pyis performed before interpolation into the prompt.
Audit Metadata