i18n-localization
Warn
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill is designed to ingest and analyze external source code using Glob, Grep, and Read. This processing of untrusted data creates an injection surface. 1. Ingestion points: Files within the user-provided <project_path> accessed via Read, Glob, and Grep. 2. Boundary markers: No delimiters or ignore-instructions markers are specified. 3. Capability inventory: Read, Glob, and Grep tools allow the agent to search and retrieve file contents. 4. Sanitization: No sanitization of ingested content is mentioned.
- [Dynamic Execution] (MEDIUM): The documentation references the execution of a Python script at 'scripts/i18n_checker.py'. Since the script's source code is not included in the skill files, its behavior and security implications cannot be verified.
Audit Metadata