NYC

long-article-to-x-content

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE] (SAFE): No malicious patterns or security risks were detected in this skill. It consists entirely of natural language templates and instructions for an AI agent to perform text summarization tasks.
  • [Indirect Prompt Injection] (SAFE): While the skill is designed to process untrusted external data (long articles provided by users), it lacks dangerous capabilities such as network access, file system writing, or command execution.
  • Ingestion points: The skill processes user-provided long articles as input for summarization.
  • Boundary markers: Absent; the prompt does not specify delimiters to separate untrusted text from instructions.
  • Capability inventory: No subprocess calls, file writes, or network operations are present.
  • Sanitization: No explicit sanitization or filtering of input text is defined.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:10 PM