maple-agent-tracing-microsoft-agent-framework

Pass

Audited by Gen Agent Trust Hub on Oct 8, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill instructs the user to install standard libraries from official registries (PyPI and NuGet), including agent-framework-core, opentelemetry-exporter-otlp-proto-http, and semantic-kernel.
  • [DATA_EXFILTRATION]: The skill configures an OTLP/HTTP exporter to send GenAI spans and conversation metadata to ingest.maple.dev. This behavior is the primary intended function of the skill (telemetry and observability) and targets the vendor's own infrastructure.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user messages and tool outputs to generate observability spans. While it captures this content (when enable_sensitive_data is set), it does so for logging purposes rather than execution, following standard observability patterns.
  • [COMMAND_EXECUTION]: Includes a manual version check command (python -c "import agent_framework; print(agent_framework.__version__)") to verify the environment before installation.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 8, 2026, 02:06 PM