xy-term-crack
Pass
Audited by Gen Agent Trust Hub on Sep 18, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions direct the agent to execute a provided Python script,
scripts/atoms-search.py, to perform keyword searches within the skill's knowledge base. - [DATA_EXFILTRATION]: The search script transmits user-provided search queries to a remote API endpoint at
https://api.xyskill.xyz/v1/atoms/search. While this is functional for the skill's retrieval mechanism, it involves sending user input to a third-party server. - [EXTERNAL_DOWNLOADS]: The skill dynamically fetches knowledge entries and scoring data (IDF) from the vendor's remote infrastructure (
api.xyskill.xyz) during the search process. - [INDIRECT_PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it ingests and processes data from external sources (remote API and local
atoms.jsonlfile). - Ingestion points: Search results from
api.xyskill.xyzand the local filereferences/atoms.jsonlare read by the script and provided to the agent. - Boundary markers: The instructions do not specify the use of delimiters or warnings to ignore instructions embedded within the retrieved knowledge items.
- Capability inventory: The skill environment allows for local file system access and network operations via the Python interpreter.
- Sanitization: There is no evidence of sanitization or validation performed on the retrieved content before it is incorporated into the agent's context.
Audit Metadata