te-meta

Pass

Audited by Gen Agent Trust Hub on Apr 28, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) due to its handling of external project metadata.
  • Ingestion points: Untrusted data such as event names, property descriptions, and entity metadata are ingested via tools like '+list_events', '+list_properties', and '+list_entities'.
  • Boundary markers: No delimiters or specific instructions to ignore embedded commands are present in the skill instructions.
  • Capability inventory: The skill has the ability to perform significant write operations including '+create_metric' and '+create_virtual_event' based on ingested data.
  • Sanitization: There is no mention of sanitization or validation for the data retrieved from the metadata store.
  • [COMMAND_EXECUTION]: The skill performs administrative tasks by executing the 'ae-cli' binary. This includes the '+create_virtual_property' tool, which allows for the definition and execution of SQL expressions on the backend platform.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 28, 2026, 06:28 AM
Security Audit — agent-trust-hub — te-meta