agento11y-test-starter
Pass
Audited by Gen Agent Trust Hub on Aug 18, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONDATA_EXFILTRATION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the
agento11yPython package and references official Grafana repositories on GitHub for Go language support. These are verified vendor resources. - [COMMAND_EXECUTION]: The skill generates a local Python runner script (
run_experiment.py) to execute agent logic and can optionally run the experiment with explicit user permission. It utilizes standard Python library mechanisms for script execution. - [DATA_EXFILTRATION]: To perform its primary function of agent observability, the skill reads project metadata (system prompts, tool definitions, and user requests) and environment configuration files (e.g.,
.env). This data, along with evaluation scores, is published to a user-configured Grafana Cloud endpoint. - [CREDENTIALS_UNSAFE]: The skill contains explicit instructions to prevent the exposure of secrets. It forbids hardcoding credentials or requesting tokens in plain text chat, instead mandating the use of environment variables and existing
.envfiles for authentication with Grafana Cloud and LLM providers.
Audit Metadata