neuropixels-analysis
Pass
Audited by Gen Agent Trust Hub on Apr 16, 2026
Risk Level: SAFE
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill recommends installing standard scientific Python packages from well-known sources, including
spikeinterface,probeinterface,kilosort, andibl-neuropixel. These are widely recognized tools in the neuroscience research community. - [COMMAND_EXECUTION]: The provided automation scripts (e.g.,
preprocess_recording.py,run_sorting.py) utilize thespikeinterfaceAPI to perform local data processing and spike sorting. These commands are consistent with the scientific purpose of the skill and do not involve suspicious or unauthorized activities. - [PROMPT_INJECTION]: The instructions focus exclusively on providing scientific guidance and tool usage. There are no attempts to override agent safety protocols or manipulate the underlying AI behavior.
- [DATA_EXFILTRATION]: No hardcoded credentials or unauthorized data transmission patterns were found. Integration with external AI models (Anthropic, OpenAI, Google) for curation tasks is performed through documented API calls that require user-provided authentication.
Audit Metadata