write
Warn
Audited by Gen Agent Trust Hub on Sep 6, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a local Python script named
emulo.pywith specific arguments to retrieve configuration paths. - [DYNAMIC_EXECUTION]: The script path is resolved dynamically at runtime by searching parent directories (
../../emulo.py), which constitutes dynamic loading from a computed path. - [INDIRECT_PROMPT_INJECTION]: The skill reads and applies instructions from external profile files returned by the
emulo.pyscript without validation. 1. Ingestion points: Profile files located at paths returned by the external script. 2. Boundary markers: None; the agent is directed to apply the context immediately. 3. Capability inventory: Ability to execute shell commands via Python. 4. Sanitization: No sanitization or integrity checks are performed on the ingested files.
Audit Metadata