repomix
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- EXTERNAL_DOWNLOADS (LOW): The skill requires installing the third-party
repomixCLI and Python utilities from public registries. This is essential for the skill's primary purpose. - PROMPT_INJECTION (LOW): The skill processes untrusted repositories for LLM consumption, creating an indirect prompt injection surface (Category 8).
- Ingestion points: Processes local and remote repositories via
repomix_batch.pyand therepomixCLI. - Boundary markers: Uses format-specific delimiters (XML, Markdown, JSON) but lacks explicit agent instructions to ignore instructions within the processed content.
- Capability inventory: Extensive file-read and concatenation capabilities to aggregate whole codebases.
- Sanitization: Integrates Secretlint to detect credentials, but lacks sanitization for malicious natural language instructions.
- COMMAND_EXECUTION (LOW): The
repomix_batch.pyscript andSKILL.mdinstructions involve executing system commands to invoke therepomixCLI, which is the intended behavior.
Audit Metadata