text-truncator
Pass
Audited by Gen Agent Trust Hub on Oct 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted text and file content, which serves as a potential vector for delivering hidden instructions to the agent.\n
- Ingestion points: Data enters through the text argument in truncate_text and the file_path argument in truncate_text_from_file within scripts/text_truncator.py.\n
- Boundary markers: The SKILL.md file does not specify the use of delimiters or protective instructions to isolate the truncated content from the agent's logic.\n
- Capability inventory: The skill possesses file system read access through the open() function in scripts/text_truncator.py.\n
- Sanitization: The implementation lacks input validation for file paths and does not perform content sanitization.
Audit Metadata