reply-writer
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources, specifically Reddit thread titles, post content, and existing comments, which could contain instructions intended to hijack the agent's output.
- Ingestion points: Step 1 (Gather Inputs) requires the user to provide the full text of an Original Post (OP) and the top existing replies.
- Boundary markers: There are no specific delimiters or instructions defined to help the agent distinguish between the user's operational parameters and the content of the Reddit thread.
- Capability inventory: The skill is limited to text generation and does not have access to sensitive files, network operations, or shell command execution.
- Sanitization: The instructions do not include steps to sanitize or filter the external content for malicious prompts or hidden instructions.
Audit Metadata