drupal-contrib-mgmt

Warn

Audited by Gen Agent Trust Hub on Mar 20, 2026

Risk Level: MEDIUMDATA_EXFILTRATIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: The skill includes instructions to generate full database dumps using drush sql:dump > backup-before-update.sql. This action creates a file containing all application data, including users and configurations, within the workspace, which could lead to unauthorized exposure if the file is not properly secured.- [COMMAND_EXECUTION]: The skill provides multiple example shell scripts that modify configuration files (cat >> composer.json), perform file system operations (rm -rf, ln -s), and execute system commands to manage dependencies, git repositories, and patches.- [EXTERNAL_DOWNLOADS]: The skill fetches content from external URLs on drupal.org, such as patch files (.patch) and RSS feeds for issue tracking. These are then used for code updates or information processing by the agent. While drupal.org is a well-known service, these downloads represent a significant attack surface.- [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection.
  • Ingestion points: Data is ingested from .info.yml files, drush upgrade_status output, and RSS feeds from drupal.org (references/issue-queue-rss-feeds.md).
  • Boundary markers: Absent. The skill provides no instructions to use delimiters or warnings when processing these external data streams.
  • Capability inventory: The skill possesses significant capabilities including file system modification, network access via curl, and package management via composer and drush.
  • Sanitization: No validation or escaping is performed on the ingested content before it is processed by the agent.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Mar 20, 2026, 05:13 PM