mistral-debug-bundle
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEDYNAMIC_CONTEXT_INJECTIONCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [DYNAMIC_CONTEXT_INJECTION]: The skill utilizes dynamic context injection markers to retrieve the version of Node.js and Python installed on the system at load time. These commands are benign system checks used for environment compatibility verification.
- [COMMAND_EXECUTION]: The diagnostic bash script executes several shell commands to gather system metadata, including 'hostname', 'uname -a', and package manager queries ('npm list', 'pip show'). It also performs a connectivity check to the official Mistral AI API endpoint.
- [INDIRECT_PROMPT_INJECTION]: The skill includes an attack surface for indirect prompt injection as it reads and processes local application logs ('logs/*.log').
- Ingestion points: Reads files in the 'logs/' directory using 'grep' to filter for error patterns.
- Boundary markers: No explicit delimiters are used for the log content, though it is tailored to the last 100 relevant lines.
- Capability inventory: The skill has access to shell execution ('Bash'), network operations ('curl'), and file archival ('tar').
- Sanitization: The script includes redaction logic ('sed') specifically for environment variables, though log content itself is not sanitized beyond filtering.
- [DYNAMIC_EXECUTION]: The skill generates a reproduction shell script ('reproduce.sh') from a static template to assist in troubleshooting. This script is intended to be run manually by the user to verify API connectivity and authentication.
Audit Metadata