Skills
skills/jeremylongshore/claude-code-plugins-plus-skills/deepgram-debug-bundle/Gen Agent Trust Hub

deepgram-debug-bundle

Pass

Audited by Gen Agent Trust Hub on Mar 12, 2026

Risk Level: SAFEDATA_EXFILTRATIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [DATA_EXFILTRATION]: The skill aggregates system metadata and log files for support purposes.
  • The script debug-env.sh captures system details using uname -a and version information for Node.js and Python environments.
  • The collect-debug-bundle.sh script extracts the last 100 lines of application logs from /var/log/app/*.log containing the string 'deepgram'. This could potentially include sensitive application data or personally identifiable information (PII) if such data is logged by the host application.
  • [COMMAND_EXECUTION]: The skill utilizes several system commands to generate diagnostic reports.
  • It executes uname, node, python3, npm list, and pip show to document the environment state.
  • The script debug-audio.sh uses ffprobe to analyze local audio file metadata.
  • [EXTERNAL_DOWNLOADS]: The skill performs network requests to verify API availability and connectivity.
  • Connectivity tests target https://api.deepgram.com/v1/projects and https://api.deepgram.com/v1/listen using curl. As Deepgram is a well-known service and these requests are for connectivity diagnostics, this interaction is considered standard for the skill's purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 12, 2026, 01:08 AM