x-reply
Pass
Audited by Gen Agent Trust Hub on Oct 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions suggest running a Python script located in a parent directory (
python3 ../x-plan/weights.py). Since this file is not part of the provided skill package, its contents and safety cannot be verified. - [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and categorize text pasted by the user from social media threads (X/Twitter replies). This external data is untrusted and could contain malicious instructions aimed at manipulating the AI agent's behavior during the drafting process.
- Ingestion points: User-pasted text or screenshots of social media replies (SKILL.md).
- Boundary markers: Absent. The skill does not define clear delimiters or instructions to ignore embedded commands within the pasted replies.
- Capability inventory: The skill generates copy-ready text drafts for the user to review. It does not possess direct file-write or network-send capabilities for this data.
- Sanitization: None. The skill processes the input directly to categorize and draft responses.
Audit Metadata