te-user-id-debug
Pass
Audited by Gen Agent Trust Hub on Aug 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to dynamically construct and execute SQL queries using the
ae-cli analysis adhoc runtool. These queries use placeholders (e.g.,[Project_ID],[Sample_aid]) that are populated based on user-provided input, which could potentially lead to SQL injection if the values are not strictly validated by the execution environment. - [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes untrusted data from external sources including SQL views (
ta.v_user_*,ta.v_event_*) and Kafka message streams. - Ingestion points: Results from SQL queries and raw Kafka message payloads (
kafka.ta."ta-data") are read into the agent's context. - Boundary markers: The instructions lack explicit boundary markers or warnings to the agent to disregard instructions potentially embedded within the ingested data.
- Capability inventory: The skill possesses the capability to execute shell-like commands via the
ae-clitool and write analysis results back to the user. - Sanitization: No explicit sanitization or filtering of the external data is specified in the workflow.
- [EXTERNAL_DOWNLOADS]: The skill provides a link to documentation on
thinkingdata.feishu.cn. Feishu is a well-known enterprise collaboration platform, and this represents a neutral reference to vendor documentation.
Audit Metadata