ag2-telemetry
Pass
Audited by Gen Agent Trust Hub on May 8, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill documentation includes instructions to install the
ag2package with tracing support via the commandpip install "ag2[openai,tracing]". - [DATA_EXFILTRATION]: The
TelemetryMiddlewareis designed to transmit execution traces to external OTLP-compatible backends such as Jaeger, Grafana Tempo, or Datadog. By default, the middleware captures and exports the content of agent messages and tool interactions. The documentation includes clear guidance on disabling this behavior using thecapture_content=Falseparameter to protect sensitive data. - [PROMPT_INJECTION]: Indirect Prompt Injection Surface: The skill ingests untrusted data from agent conversations and tool outputs to populate telemetry spans. This establishes a surface where malicious content processed by the agent is included in logs sent to external monitoring systems.
Audit Metadata