actualizar-wachines

Pass

Audited by Gen Agent Trust Hub on Sep 6, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill fetches plugin code and marketplace metadata from a remote GitHub repository: https://github.com/Perennia-Regeneracion/Wachines-Plugin-Ironman.git. This repository serves as the source for the Wachines-Plugin-Ironman plugin updates.
  • [COMMAND_EXECUTION]: The skill executes shell commands to interact with the platform's plugin manager, including claude plugin marketplace update wachines and claude plugin marketplace list. These commands perform git operations and display installed plugin information.
  • [INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection as it ingests and installs plugin code from an external source that could contain instructions influencing agent behavior.
  • Ingestion points: Fetches data from the Perennia-Regeneracion GitHub organization via the claude plugin utility.
  • Boundary markers: None identified; the skill assumes the remote source is authoritative and safe for update.
  • Capability inventory: Executes shell commands and interacts with the agent's plugin loading system (/reload-plugins).
  • Sanitization: Relies on the agent platform's internal plugin loading mechanisms without additional validation of the fetched content.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 6, 2026, 04:39 PM
Security Audit — agent-trust-hub — actualizar-wachines