maple-agent-tracing-agno
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill follows security best practices by recommending that sensitive API keys be managed via environment variables or secret managers rather than being hardcoded. It correctly identifies the difference between private and public ingest keys and advises against exposing private keys in browser-side code.
- [COMMAND_EXECUTION]: The skill provides instructions for installing well-known Python packages ('agno', 'opentelemetry-sdk', 'openinference-instrumentation-agno') using standard package managers. These operations are within the scope of the skill's primary purpose of setting up observability.
- [INDIRECT_PROMPT_INJECTION]: The skill exhibits an attack surface by reading and modifying project configuration and source code files. However, this surface is essential for its intended functionality of automating instrumentation.
- Ingestion points: Project configuration files (e.g., 'pyproject.toml', 'requirements.txt') and Python entry point files (e.g., 'main.py', 'app.py').
- Boundary markers: Not explicitly defined in the skill, relying on standard agent delimiters.
- Capability inventory: CLI-based package installation, file creation ('tracing.py'), and source code modification.
- Sanitization: The skill includes guidance on masking PII and sensitive messages (e.g., via 'OPENINFERENCE_HIDE_INPUT_MESSAGES') if the user has specific privacy requirements.
Audit Metadata