content-repurposing

Fail

Audited by Gen Agent Trust Hub on Feb 18, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • Remote Code Execution (HIGH): The skill instructs the agent to run curl -fsSL https://cli.inference.sh | sh. This is a dangerous pattern that executes arbitrary code from an untrusted domain without verification or integrity checks.\n- External Downloads (MEDIUM): The skill uses npx skills add to dynamically load multiple external packages from the inference-sh repository. These sources are not included in the trusted whitelist, posing a supply chain risk.\n- Command Execution (LOW): The skill requires Bash(infsh *) privileges to run various AI application tools, granting the agent broad execution capabilities within that CLI namespace.\n- Prompt Injection (LOW): The skill is highly vulnerable to Indirect Prompt Injection (Category 8) because it is designed to ingest and process long-form untrusted content like blog posts and podcast transcripts.\n
  • Ingestion points: Processes external text and audio data via the infsh CLI.\n
  • Boundary markers: None present in the instructions to separate data from commands.\n
  • Capability inventory: Full Bash access (scoped to infsh) and network operations.\n
  • Sanitization: No input validation or filtering is performed on the ingested content.
Recommendations
  • HIGH: Downloads and executes remote code from: https://cli.inference.sh - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Feb 18, 2026, 01:35 AM
Security Audit — agent-trust-hub — content-repurposing