ai-tracking-experiments

Pass

Audited by Gen Agent Trust Hub on May 13, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill uses standard file operations to manage local experiment logs (JSONL) and artifacts, which is consistent with its stated purpose of experiment tracking.
  • [EXTERNAL_DOWNLOADS]: Recommends installing well-known packages (weave, langwatch) from official registries for monitoring and collaboration. These originate from established technology providers.
  • [COMMAND_EXECUTION]: Uses routine shell commands for package installation (pip install) and adding related skills from the same author (npx skills add), all of which are documented and expected for setup.
  • [PROMPT_INJECTION]: No instructions aimed at overriding agent behavior or bypassing safety filters were detected. The language is purely instructional and task-oriented.
  • [DATA_EXFILTRATION]: No evidence of sensitive data harvesting or unauthorized network transmissions was found. Network activity is limited to official API integrations with W&B and LangWatch.
Audit Metadata
Risk Level
SAFE
Analyzed
May 13, 2026, 06:45 PM