cloudflare-access-troubleshoot

Warn

Audited by Gen Agent Trust Hub on Feb 24, 2026

Risk Level: MEDIUMCREDENTIALS_UNSAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [CREDENTIALS_UNSAFE]: The skill uses grep to extract sensitive credentials including GOOGLE_OAUTH_CLIENT_ID, GOOGLE_OAUTH_CLIENT_SECRET, and ACCESS_ALLOWED_EMAIL from the local .env file.
  • [COMMAND_EXECUTION]: Includes commands requiring elevated privileges (sudo) to flush DNS caches (dscacheutil, systemd-resolve) on macOS and Linux.
  • [COMMAND_EXECUTION]: Executes several local shell scripts located in the author's project directory (/home/dawiddutoit/projects/network/scripts/) to modify network configurations and Access policies.
  • [COMMAND_EXECUTION]: Uses docker exec to query and potentially modify the state of a Pi-hole container.
  • [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection by reading external data from configuration files.
  • Ingestion points: Reads /home/dawiddutoit/projects/network/.env using grep and nano.
  • Boundary markers: None identified in the prompt templates.
  • Capability inventory: Full Bash access, script execution, Docker commands, and file read/write capabilities.
  • Sanitization: No validation or sanitization of data retrieved from the configuration files before potential use in subsequent logic.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 24, 2026, 05:23 PM