write-aida-article-create

Fail

Audited by Gen Agent Trust Hub on Jun 18, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill downloads content from a remote server (https://jeffbailey.us/prompts/aida-article-create/raw.html) and instructs the agent to follow the instructions contained within that file. This mechanism for dynamic instruction loading from an untrusted external source allows for remote modification of the agent's behavior.
  • [COMMAND_EXECUTION]: The skill provides and expects the execution of shell commands, including mkdir, curl, and pandoc, to manage a local prompt cache. This involves direct interaction with the host's file system and network.
  • [EXTERNAL_DOWNLOADS]: Fetches data and instruction templates from a remote domain (jeffbailey.us) that is not on the trusted vendors list. While the domain is similar to the author's name, the character discrepancy (jeffbailey vs jeffabailey) requires it to be treated as an untrusted source.
Recommendations
  • HIGH: Downloads and executes remote code from: https://jeffbailey.us/prompts/aida-article-create/raw.html - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Jun 18, 2026, 10:47 AM